Trying to restore a level 0 backup to a different,
A DBA testing disaster recovery on HP-UX 11.11 / IDS 9.40.FC3 tried 'ontape -r' of a level-0 archive onto a supposedly identical standby box and got "Physical restore failed - could not fork server connection, SQLCODE -931". Replies identified -931 as the sqlhosts servicename missing from /etc/services on the target; adding the sqlexec entry cleared that error, but the restore then failed with "oninit: Fatal error in shared memory creation". Further advice was to keep INFORMIXSERVER, ONCONFIG (incl. file name), sqlhosts and all chunk paths/sizes/offsets consistent, and to check the MSGPATH log. No confirmation of a final fix is recorded in the thread.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Server: HP/UX 11.11
Informix: 9.4FC3
We are attempting to 'test' our backup solution and received the
following error when doing an 'ontape -r' on the backup server.
Physical restore failed - could not fork server connection, SQLCODE -931
We have the tapedev configured as a file on a remote server
(/remote/recovery/informix/level0) on both servers.
The hardware and chunk links are identical on both servers.
The onconfig & host has different server names so when I bring up the
backup (if I can restore the backup), they won't have same names.
It is a single instance with 24 dbspaces & 7 non system databases.
The backups have run without a problem for a couple of years and have
never needed to test recovery (and didn't have the downtime to really
try). Now we have this identical server and want to ensure our backup
solution actually works. Still not sure:)
I have read through the Backup & Restore guide several times for Ontape
restores and it doesn't say a whole lot, run ontape -r and you should be
good to go.
I pressed enter at the mount tape prompt, entered yes to continue at
next prompt, no to backup logs on disk, it ran for about 30 minutes then
I noticed the error message. Below is the complete attempt. Any help
would be greatly appreciated.
Thanks,
Randy K.
=>ontape -r
Please mount tape 1 on /remote/recovery/informix/level0 and press Return
to continue ...
Archive Tape Information
Tape type: Archive Backup Tape
Online version: IBM Informix Dynamic Server Version 9.40.FC3
Archive date: Sun Mar 9 18:39:29 2008
User id: informix
Terminal id: /dev/console
Archive level: 0
Tape device: /remote/recovery/informix/level0
Tape blocksize (in k): 16
Tape size (in k): 30000000
Tape number in series: 1
Spaces to restore:1 [rootdbs
]
2 [llogdbs
]
3 [casedbs
]
4 [physdbs
]
5 [chrgdbs
]
6 [dwdbs
]
7 [clndrdbs
]
8 [ctntndbs
]
9 [eventdbs
]
10 [masterdbs
]
11 [peopledbs
]
12 [smat2dbs
]
13 [awdbs
]
14 [rcptmstrdbs
]
15 [sccdbs
]
16 [cstrkdbs
]
17 [testdbs
]
Archive Information
IBM Informix Dynamic Server Copyright(C) 1986-1999 IBM Informix
Software, Inc.
Initialization Time 04/21/2004 12:46:09
System Page Size 2048
Version 13
Archive CheckPoint Time 03/09/2008 18:39:29
Dbspaces
number flags fchunk nchunks flags owner
name
1 40001 1 1 N B informix
rootdbs
2 40001 2 1 N B informix
llogdbs
3 40001 3 1 N B informix
casedbs
4 40001 4 1 N B informix
physdbs
5 42001 5 1 N TB informix
tempdbs1
6 42001 6 1 N TB informix
tempdbs2
7 42001 7 1 N TB informix
tempdbs3
8 42001 8 1 N TB informix
tempdbs4
9 40001 9 1 N B informix
chrgdbs
10 40001 10 1 N B informix
dwdbs
11 40001 11 1 N B informix
clndrdbs
12 40001 12 1 N B informix
ctntndbs
13 40001 13 3 N B informix
eventdbs
14 40001 14 2 N B informix
masterdbs
15 40001 15 1 N B informix
peopledbs
17 40001 17 1 N B informix
smat2dbs
18 40001 18 1 N B informix
awdbs
19 40001 19 1 N B informix
rcptmstrdbs
20 40001 22 1 N B informix
sccdbs
21 40001 23 1 N B informix
cstrkdbs
22 40001 24 1 N B informix
testdbs
Chunks
chk/dbs offset size free bpages flags pathname
chk/dbs offset size free bpages flags pathname
1 1 0 192000 158948 PO-B /db/links/rootdbs
2 2 0 512000 1947 PO-B /db/links/llogdbs
3 3 0 256000 54780 PO-B /db/links/casedbs
4 4 0 36000 35947 PO-B /db/links/physdbs
5 5 0 512000 511847 PO-B /db/links/tempdbs1
6 6 0 512000 511847 PO-B /db/links/tempdbs2
7 7 0 512000 511839 PO-B /db/links/tempdbs3
8 8 0 512000 511839 PO-B /db/links/tempdbs4
9 9 0 512000 194092 PO-B /db/links/chrgdbs
10 10 0 1024000 752727 PO-B /db/links/dwdbs
11 11 0 1536000 958131 PO-B /db/links/clndrdbs
12 12 0 512000 286427 PO-B /db/links/ctntndbs
13 13 0 2048000 10188 PO-B /db/links/eventdbs
14 14 0 2048000 388648 PO-B /db/links/masterdbs
15 15 0 1536000 524717 PO-B /db/links/peopledbs
16 13 0 512000 229413 PO-B /db/links/1gbdbs3
17 17 0 6144000 6138738 PO-B /db/links/smat2dbs1
18 18 0 1048576 974846 PO-B /db/links/awdbs
19 19 0 512000 88812 PO-B /db/links/rcptmstrdbs
20 14 0 512000 511997 PO-B /db/links/1gbdbs1
21 13 0 1048576 0 PO-B /db/links/eventdbs2
22 20 0 1048576 871219 PO-B /db/links/sccdbs
23 21 0 512000 465169 PO-B /db/links/1gbdbs2
24 22 0 512000 325544 PO-B /db/links/testdbs
Continue restore? (y/n)y
Do you want to back up the logs? (y/n)n
Physical restore failed - could not fork server connection, SQLCODE -931
Kennedy, Randy wrote:
> Server: HP/UX 11.11
> Informix: 9.4FC3
>
> We are attempting to 'test' our backup solution and received the
> following error when doing an 'ontape -r' on the backup server.
> Physical restore failed - could not fork server connection, SQLCODE -931
>
> We have the tapedev configured as a file on a remote server
> (/remote/recovery/informix/level0) on both servers.
> The hardware and chunk links are identical on both servers.
> The onconfig & host has different server names so when I bring up the
> backup (if I can restore the backup), they won't have same names.
> It is a single instance with 24 dbspaces & 7 non system databases.
> The backups have run without a problem for a couple of years and have
> never needed to test recovery (and didn't have the downtime to really
> try). Now we have this identical server and want to ensure our backup
> solution actually works. Still not sure:)
>
> I have read through the Backup & Restore guide several times for Ontape
> restores and it doesn't say a whole lot, run ontape -r and you should be
> good to go.
>
> I pressed enter at the mount tape prompt, entered yes to continue at
> next prompt, no to backup logs on disk, it ran for about 30 minutes then
> I noticed the error message. Below is the complete attempt. Any help
> would be greatly appreciated.
This is the clue:
> Physical restore failed - could not fork server connection, SQLCODE -931
>
Finderr -931 would report that the servicename for the server in the
sqlhosts file is not found in /etc/services.
Art S. Kagel
Oninit
================================================================================
===========
Please access the attached hyperlink for an important electronic
communications disclaimer:
http://www.oninit.com/home/disclaimer.php
================================================================================
===========
Randy,
Have you taken a look at your configuration files for both the instance=
and
the Unix host?
The -931 error indicates that the servicename is not found in
/etc/services. You should
take a look at the $INFORMIXDIR/etc/sqlhosts or $INFORMIXSQLHOSTS file =
and
see
what you have set for the servicename. While you are at it, you should=
insure that the other
values are valid as well to include the hostname.
Good luck.
=
"Kennedy, Randy" =
<RKennedy@scottsd =
aleaz.gov> =
To
Sent by: ids@iiug.org =
ids-bounces@iiug. =
cc
org =
Subj=
ect
Trying to restore a level 0 back=
up
03/10/2008 04:27 to a differ.... [11559] =
PM =
=
=
Please respond to =
ids@iiug.org =
=
=
Server: HP/UX 11.11
Informix: 9.4FC3
We are attempting to 'test' our backup solution and received the
following error when doing an 'ontape -r' on the backup server.
Physical restore failed - could not fork server connection, SQLCODE -93=
1
We have the tapedev configured as a file on a remote server
(/remote/recovery/informix/level0) on both servers.
The hardware and chunk links are identical on both servers.
The onconfig & host has different server names so when I bring up the
backup (if I can restore the backup), they won't have same names.
It is a single instance with 24 dbspaces & 7 non system databases.
The backups have run without a problem for a couple of years and have
never needed to test recovery (and didn't have the downtime to really
try). Now we have this identical server and want to ensure our backup
solution actually works. Still not sure:)
I have read through the Backup & Restore guide several times for Ontape=
restores and it doesn't say a whole lot, run ontape -r and you should b=
e
good to go.
I pressed enter at the mount tape prompt, entered yes to continue at
next prompt, no to backup logs on disk, it ran for about 30 minutes the=
n
I noticed the error message. Below is the complete attempt. Any help
would be greatly appreciated.
Thanks,
Randy K.
=3D>ontape -r
Please mount tape 1 on /remote/recovery/informix/level0 and press Retur=
n
to continue ...
Archive Tape Information
Tape type: Archive Backup Tape
Online version: IBM Informix Dynamic Server Version 9.40.FC3
Archive date: Sun Mar 9 18:39:29 2008
User id: informix
Terminal id: /dev/console
Archive level: 0
Tape device: /remote/recovery/informix/level0
Tape blocksize (in k): 16
Tape size (in k): 30000000
Tape number in series: 1
Spaces to restore:1 [rootdbs
]
2 [llogdbs
]
3 [casedbs
]
4 [physdbs
]
5 [chrgdbs
]
6 [dwdbs
]
7 [clndrdbs
]
8 [ctntndbs
]
9 [eventdbs
]
10 [masterdbs
]
11 [peopledbs
]
12 [smat2dbs
]
13 [awdbs
]
14 [rcptmstrdbs
]
15 [sccdbs
]
16 [cstrkdbs
]
17 [testdbs
]
Archive Information
IBM Informix Dynamic Server Copyright(C) 1986-1999 IBM Informix
Software, Inc.
Initialization Time 04/21/2004 12:46:09
System Page Size 2048
Version 13
Archive CheckPoint Time 03/09/2008 18:39:29
Dbspaces
number flags fchunk nchunks flags owner
name
1 40001 1 1 N B informix
rootdbs
2 40001 2 1 N B informix
llogdbs
3 40001 3 1 N B informix
casedbs
4 40001 4 1 N B informix
physdbs
5 42001 5 1 N TB informix
tempdbs1
6 42001 6 1 N TB informix
tempdbs2
7 42001 7 1 N TB informix
tempdbs3
8 42001 8 1 N TB informix
tempdbs4
9 40001 9 1 N B informix
chrgdbs
10 40001 10 1 N B informix
dwdbs
11 40001 11 1 N B informix
clndrdbs
12 40001 12 1 N B informix
ctntndbs
13 40001 13 3 N B informix
eventdbs
14 40001 14 2 N B informix
masterdbs
15 40001 15 1 N B informix
peopledbs
17 40001 17 1 N B informix
smat2dbs
18 40001 18 1 N B informix
awdbs
19 40001 19 1 N B informix
rcptmstrdbs
20 40001 22 1 N B informix
sccdbs
21 40001 23 1 N B informix
cstrkdbs
22 40001 24 1 N B informix
testdbs
Chunks
chk/dbs offset size free bpages flags pathname
chk/dbs offset size free bpages flags pathname
1 1 0 192000 158948 PO-B /db/links/rootdbs
2 2 0 512000 1947 PO-B /db/links/llogdbs
3 3 0 256000 54780 PO-B /db/links/casedbs
4 4 0 36000 35947 PO-B /db/links/physdbs
5 5 0 512000 511847 PO-B /db/links/tempdbs1
6 6 0 512000 511847 PO-B /db/links/tempdbs2
7 7 0 512000 511839 PO-B /db/links/tempdbs3
8 8 0 512000 511839 PO-B /db/links/tempdbs4
9 9 0 512000 194092 PO-B /db/links/chrgdbs
10 10 0 1024000 752727 PO-B /db/links/dwdbs
11 11 0 1536000 958131 PO-B /db/links/clndrdbs
12 12 0 512000 286427 PO-B /db/links/ctntndbs
13 13 0 2048000 10188 PO-B /db/links/eventdbs
14 14 0 2048000 388648 PO-B /db/links/masterdbs
15 15 0 1536000 524717 PO-B /db/links/peopledbs
16 13 0 512000 229413 PO-B /db/links/1gbdbs3
17 17 0 6144000 6138738 PO-B /db/links/smat2dbs1
18 18 0 1048576 974846 PO-B /db/links/awdbs
19 19 0 512000 88812 PO-B /db/links/rcptmstrdbs
20 14 0 512000 511997 PO-B /db/links/1gbdbs1
21 13 0 1048576 0 PO-B /db/links/eventdbs2
22 20 0 1048576 871219 PO-B /db/links/sccdbs
23 21 0 512000 465169 PO-B /db/links/1gbdbs2
24 22 0 512000 325544 PO-B /db/links/testdbs
Continue restore? (y/n)y
Do you want to back up the logs? (y/n)n
Physical restore failed - could not fork server connection, SQLCODE -93=
1
***********************************************************************=
********
Forum Note: Use "Reply" to post a response in the discussion forum.
See you at the IIUG Informix 2008 Conference
The Power Conference for Informix Professionals
April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
http://www.iiug.org/conf
Registration Now Open!!
=
So much for it being an identical configuration as I was told by the
sysadmin who prepared it:)
The sqlexec entry was indeed missing from /etc/services.
I added it and tried to rerun ontape -r and got this message (and the
prompt didn't return):
oninit: Fatal error in shared memory creation
Some one else stated to look at $INFORMIXDIR/etc/sqlhosts. On the
system that the backup was created on is courtshm and I changed that to
courtbak on the system I am trying to restore it to. This matches the
name I changed servername to in onconfig to avoid the duplicate when
restored.
Thanks,
Randy
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
Art S. Kagel (Oninit)
Sent: Monday, March 10, 2008 1:43 PM
To: ids@iiug.org
Subject: Re: Trying to restore a level 0 backup to a di.... [11560]
Kennedy, Randy wrote:
> Server: HP/UX 11.11
> Informix: 9.4FC3
>
> We are attempting to 'test' our backup solution and received the
> following error when doing an 'ontape -r' on the backup server.
> Physical restore failed - could not fork server connection, SQLCODE
> -931
>
> We have the tapedev configured as a file on a remote server
> (/remote/recovery/informix/level0) on both servers.
> The hardware and chunk links are identical on both servers.
> The onconfig & host has different server names so when I bring up the
> backup (if I can restore the backup), they won't have same names.
> It is a single instance with 24 dbspaces & 7 non system databases.
> The backups have run without a problem for a couple of years and have
> never needed to test recovery (and didn't have the downtime to really
> try). Now we have this identical server and want to ensure our backup
> solution actually works. Still not sure:)
>
> I have read through the Backup & Restore guide several times for
> Ontape restores and it doesn't say a whole lot, run ontape -r and you
> should be good to go.
>
> I pressed enter at the mount tape prompt, entered yes to continue at
> next prompt, no to backup logs on disk, it ran for about 30 minutes
> then I noticed the error message. Below is the complete attempt. Any
> help would be greatly appreciated.
This is the clue:
> Physical restore failed - could not fork server connection, SQLCODE
> -931
>
Finderr -931 would report that the servicename for the server in the
sqlhosts file is not found in /etc/services.
Art S. Kagel
Oninit
========================================================================
===================
Please access the attached hyperlink for an important electronic
communications disclaimer:
http://www.oninit.com/home/disclaimer.php
========================================================================
===================
************************************************************************
*******
Forum Note: Use "Reply" to post a response in the discussion forum.
See you at the IIUG Informix 2008 Conference
The Power Conference for Informix Professionals
April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
http://www.iiug.org/conf
Registration Now Open!!
Mkae sure all chunks are present, have the same name., same size and same
offset.
Make sure that the onconfig files and sqlhost files are the same.
One way of checking is to create a new instance, using the full onconfig file
and creating all dbspaces that will be available from the original instance.
If you get this to work, then restoring an archive to a new system should work.
"Kennedy, Randy" <RKennedy@scottsdaleaz.gov> wrote:
Server: HP/UX 11.11
Informix: 9.4FC3
We are attempting to 'test' our backup solution and received the
following error when doing an 'ontape -r' on the backup server.
Physical restore failed - could not fork server connection, SQLCODE -931
We have the tapedev configured as a file on a remote server
(/remote/recovery/informix/level0) on both servers.
The hardware and chunk links are identical on both servers.
The onconfig & host has different server names so when I bring up the
backup (if I can restore the backup), they won't have same names.
It is a single instance with 24 dbspaces & 7 non system databases.
The backups have run without a problem for a couple of years and have
never needed to test recovery (and didn't have the downtime to really
try). Now we have this identical server and want to ensure our backup
solution actually works. Still not sure:)
I have read through the Backup & Restore guide several times for Ontape
restores and it doesn't say a whole lot, run ontape -r and you should be
good to go.
I pressed enter at the mount tape prompt, entered yes to continue at
next prompt, no to backup logs on disk, it ran for about 30 minutes then
I noticed the error message. Below is the complete attempt. Any help
would be greatly appreciated.
Thanks,
Randy K.
=>ontape -r
Please mount tape 1 on /remote/recovery/informix/level0 and press Return
to continue ...
Archive Tape Information
Tape type: Archive Backup Tape
Online version: IBM Informix Dynamic Server Version 9.40.FC3
Archive date: Sun Mar 9 18:39:29 2008
User id: informix
Terminal id: /dev/console
Archive level: 0
Tape device: /remote/recovery/informix/level0
Tape blocksize (in k): 16
Tape size (in k): 30000000
Tape number in series: 1
Spaces to restore:1 [rootdbs
]
2 [llogdbs
]
3 [casedbs
]
4 [physdbs
]
5 [chrgdbs
]
6 [dwdbs
]
7 [clndrdbs
]
8 [ctntndbs
]
9 [eventdbs
]
10 [masterdbs
]
11 [peopledbs
]
12 [smat2dbs
]
13 [awdbs
]
14 [rcptmstrdbs
]
15 [sccdbs
]
16 [cstrkdbs
]
17 [testdbs
]
Archive Information
IBM Informix Dynamic Server Copyright(C) 1986-1999 IBM Informix
Software, Inc.
Initialization Time 04/21/2004 12:46:09
System Page Size 2048
Version 13
Archive CheckPoint Time 03/09/2008 18:39:29
Dbspaces
number flags fchunk nchunks flags owner
name
1 40001 1 1 N B informix
rootdbs
2 40001 2 1 N B informix
llogdbs
3 40001 3 1 N B informix
casedbs
4 40001 4 1 N B informix
physdbs
5 42001 5 1 N TB informix
tempdbs1
6 42001 6 1 N TB informix
tempdbs2
7 42001 7 1 N TB informix
tempdbs3
8 42001 8 1 N TB informix
tempdbs4
9 40001 9 1 N B informix
chrgdbs
10 40001 10 1 N B informix
dwdbs
11 40001 11 1 N B informix
clndrdbs
12 40001 12 1 N B informix
ctntndbs
13 40001 13 3 N B informix
eventdbs
14 40001 14 2 N B informix
masterdbs
15 40001 15 1 N B informix
peopledbs
17 40001 17 1 N B informix
smat2dbs
18 40001 18 1 N B informix
awdbs
19 40001 19 1 N B informix
rcptmstrdbs
20 40001 22 1 N B informix
sccdbs
21 40001 23 1 N B informix
cstrkdbs
22 40001 24 1 N B informix
testdbs
Chunks
chk/dbs offset size free bpages flags pathname
chk/dbs offset size free bpages flags pathname
1 1 0 192000 158948 PO-B /db/links/rootdbs
2 2 0 512000 1947 PO-B /db/links/llogdbs
3 3 0 256000 54780 PO-B /db/links/casedbs
4 4 0 36000 35947 PO-B /db/links/physdbs
5 5 0 512000 511847 PO-B /db/links/tempdbs1
6 6 0 512000 511847 PO-B /db/links/tempdbs2
7 7 0 512000 511839 PO-B /db/links/tempdbs3
8 8 0 512000 511839 PO-B /db/links/tempdbs4
9 9 0 512000 194092 PO-B /db/links/chrgdbs
10 10 0 1024000 752727 PO-B /db/links/dwdbs
11 11 0 1536000 958131 PO-B /db/links/clndrdbs
12 12 0 512000 286427 PO-B /db/links/ctntndbs
13 13 0 2048000 10188 PO-B /db/links/eventdbs
14 14 0 2048000 388648 PO-B /db/links/masterdbs
15 15 0 1536000 524717 PO-B /db/links/peopledbs
16 13 0 512000 229413 PO-B /db/links/1gbdbs3
17 17 0 6144000 6138738 PO-B /db/links/smat2dbs1
18 18 0 1048576 974846 PO-B /db/links/awdbs
19 19 0 512000 88812 PO-B /db/links/rcptmstrdbs
20 14 0 512000 511997 PO-B /db/links/1gbdbs1
21 13 0 1048576 0 PO-B /db/links/eventdbs2
22 20 0 1048576 871219 PO-B /db/links/sccdbs
23 21 0 512000 465169 PO-B /db/links/1gbdbs2
24 22 0 512000 325544 PO-B /db/links/testdbs
Continue restore? (y/n)y
Do you want to back up the logs? (y/n)n
Physical restore failed - could not fork server connection, SQLCODE -931
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
See you at the IIUG Informix 2008 Conference
The Power Conference for Informix Professionals
April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
http://www.iiug.org/conf
Registration Now Open!!
---------------------------------
Looking for last minute shopping deals? Find them fast with Yahoo! Search.
Hi,
a) when changing the name of the server, also change the environment
variable INFORMIXSERVER accordingly. (There are 3 places: the
environment variable, the entry in onconfig file and the entry in
sqlhosts
file.)
b) check the message log (see MSGPATH in your onconfig file).
It may contain useful information to determine the cause of the
problem.
Regards,
Martin
--
Martin Fuerderer
IBM Informix Development Munich, Germany
Information Management
IBM Deutschland Entwicklung GmbH
Chairman of the Supervisory Board: Martin Jetter
Board of Management: Herbert Kircher
Corporate Seat: Boeblingen, Germany
Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294
ids-bounces@iiug.org wrote on 10.03.2008 23:11:22:
> So much for it being an identical configuration as I was told by the
> sysadmin who prepared it:)
>
> The sqlexec entry was indeed missing from /etc/services.
>
> I added it and tried to rerun ontape -r and got this message (and the
> prompt didn't return):
>
> oninit: Fatal error in shared memory creation
>
> Some one else stated to look at $INFORMIXDIR/etc/sqlhosts. On the
> system that the backup was created on is courtshm and I changed that to
> courtbak on the system I am trying to restore it to. This matches the
> name I changed servername to in onconfig to avoid the duplicate when
> restored.
>
> Thanks,
> Randy
>
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> Art S. Kagel (Oninit)
> Sent: Monday, March 10, 2008 1:43 PM
> To: ids@iiug.org
> Subject: Re: Trying to restore a level 0 backup to a di.... [11560]
>
> Kennedy, Randy wrote:
> > Server: HP/UX 11.11
> > Informix: 9.4FC3
> >
> > We are attempting to 'test' our backup solution and received the
> > following error when doing an 'ontape -r' on the backup server.
> > Physical restore failed - could not fork server connection, SQLCODE
> > -931
> >
> > We have the tapedev configured as a file on a remote server
> > (/remote/recovery/informix/level0) on both servers.
> > The hardware and chunk links are identical on both servers.
> > The onconfig & host has different server names so when I bring up the
> > backup (if I can restore the backup), they won't have same names.
> > It is a single instance with 24 dbspaces & 7 non system databases.
> > The backups have run without a problem for a couple of years and have
> > never needed to test recovery (and didn't have the downtime to really
> > try). Now we have this identical server and want to ensure our backup
> > solution actually works. Still not sure:)
> >
> > I have read through the Backup & Restore guide several times for
> > Ontape restores and it doesn't say a whole lot, run ontape -r and you
> > should be good to go.
> >
> > I pressed enter at the mount tape prompt, entered yes to continue at
> > next prompt, no to backup logs on disk, it ran for about 30 minutes
> > then I noticed the error message. Below is the complete attempt. Any
> > help would be greatly appreciated.
>
> This is the clue:
> > Physical restore failed - could not fork server connection, SQLCODE
> > -931
> >
>
> Finderr -931 would report that the servicename for the server in the
> sqlhosts file is not found in /etc/services.
>
> Art S. Kagel
> Oninit
>
> ========================================================================
> ===================
> Please access the attached hyperlink for an important electronic
> communications disclaimer:
>
> http://www.oninit.com/home/disclaimer.php
>
> ========================================================================
> ===================
>
> ************************************************************************
> *******
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
> See you at the IIUG Informix 2008 Conference
> The Power Conference for Informix Professionals
> April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> http://www.iiug.org/conf
> Registration Now Open!!
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
> See you at the IIUG Informix 2008 Conference
> The Power Conference for Informix Professionals
> April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> http://www.iiug.org/conf
> Registration Now Open!!
Hi Martin,
I think you have to change the name of the onconfig file, also.
Regards
Knuth
On Tue, 2008-03-11 at 04:38 -0400, Martin Fuerderer wrote:
> Hi,
>
> a) when changing the name of the server, also change the environment
> variable INFORMIXSERVER accordingly. (There are 3 places: the
> environment variable, the entry in onconfig file and the entry in
> sqlhosts
> file.)
>
> b) check the message log (see MSGPATH in your onconfig file).
> It may contain useful information to determine the cause of the
> problem.
>
> Regards,
> Martin
> --
> Martin Fuerderer
> IBM Informix Development Munich, Germany
> Information Management
>
> IBM Deutschland Entwicklung GmbH
> Chairman of the Supervisory Board: Martin Jetter
> Board of Management: Herbert Kircher
> Corporate Seat: Boeblingen, Germany
> Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294
>
> ids-bounces@iiug.org wrote on 10.03.2008 23:11:22:
>
> > So much for it being an identical configuration as I was told by the
> > sysadmin who prepared it:)
> >
> > The sqlexec entry was indeed missing from /etc/services.
> >
> > I added it and tried to rerun ontape -r and got this message (and the
> > prompt didn't return):
> >
> > oninit: Fatal error in shared memory creation
> >
> > Some one else stated to look at $INFORMIXDIR/etc/sqlhosts. On the
> > system that the backup was created on is courtshm and I changed that to
> > courtbak on the system I am trying to restore it to. This matches the
> > name I changed servername to in onconfig to avoid the duplicate when
> > restored.
> >
> > Thanks,
> > Randy
> >
> > -----Original Message-----
> > From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> > Art S. Kagel (Oninit)
> > Sent: Monday, March 10, 2008 1:43 PM
> > To: ids@iiug.org
> > Subject: Re: Trying to restore a level 0 backup to a di.... [11560]
> >
> > Kennedy, Randy wrote:
> > > Server: HP/UX 11.11
> > > Informix: 9.4FC3
> > >
> > > We are attempting to 'test' our backup solution and received the
> > > following error when doing an 'ontape -r' on the backup server.
> > > Physical restore failed - could not fork server connection, SQLCODE
> > > -931
> > >
> > > We have the tapedev configured as a file on a remote server
> > > (/remote/recovery/informix/level0) on both servers.
> > > The hardware and chunk links are identical on both servers.
> > > The onconfig & host has different server names so when I bring up the
> > > backup (if I can restore the backup), they won't have same names.
> > > It is a single instance with 24 dbspaces & 7 non system databases.
> > > The backups have run without a problem for a couple of years and have
> > > never needed to test recovery (and didn't have the downtime to really
> > > try). Now we have this identical server and want to ensure our backup
> > > solution actually works. Still not sure:)
> > >
> > > I have read through the Backup & Restore guide several times for
> > > Ontape restores and it doesn't say a whole lot, run ontape -r and you
> > > should be good to go.
> > >
> > > I pressed enter at the mount tape prompt, entered yes to continue at
> > > next prompt, no to backup logs on disk, it ran for about 30 minutes
> > > then I noticed the error message. Below is the complete attempt. Any
> > > help would be greatly appreciated.
> >
> > This is the clue:
> > > Physical restore failed - could not fork server connection, SQLCODE
> > > -931
> > >
> >
> > Finderr -931 would report that the servicename for the server in the
> > sqlhosts file is not found in /etc/services.
> >
> > Art S. Kagel
> > Oninit
> >
> > ========================================================================
>
> > ===================
> > Please access the attached hyperlink for an important electronic
> > communications disclaimer:
> >
> > http://www.oninit.com/home/disclaimer.php
> >
> > ========================================================================
>
> > ===================
> >
> > ************************************************************************
>
> > *******
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
> > See you at the IIUG Informix 2008 Conference
> > The Power Conference for Informix Professionals
> > April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> > http://www.iiug.org/conf
> > Registration Now Open!!
> >
> >
> >
>
>
*******************************************************************************
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
> > See you at the IIUG Informix 2008 Conference
> > The Power Conference for Informix Professionals
> > April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> > http://www.iiug.org/conf
> > Registration Now Open!!
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
> See you at the IIUG Informix 2008 Conference
> The Power Conference for Informix Professionals
> April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> http://www.iiug.org/conf
> Registration Now Open!!
Knuth Hartlieb
IMPIRE AG
"AGROB-Medienpark" Geb. 5
Münchener Straße 101B
D-85737 Ismaning
Phone: +49 (0) 89-4 52 25-4 61
Fax: +49 (0) 89-4 52 25-2 90
Mobil: +49 (0) 1 72-4 38 35 00
E-mail: knuth.hartlieb@impire.de
Skype: kjhartlieb
ITToolbox: kjhartlieb
Besuchen Sie uns unter
www.impire.de & www.bundesliga-datenbank.de
IMPIRE AG: Sitz: D-85737 Ismaning, Amtsgericht München HR B 152 835;
Vorstand: Hartmut Braun (Vorsitzender), Harald Keilbach;
Aufsichtsratsvorsitzender: Bernd Künzler
CAIROS technologies AG; Sitz: 76307 Karlsbad, Amtsgericht Mannheim HRB
362651; Vorstand: Hartmut Braun (Vorsitzender), Christian Holzer;
Aufsichtsratsvorsitzender: Roland Schmider
Die in dieser E-Mail beinhaltete Information oder Anlagen zu dieser E-
Mail (im Folgenden Mitteilung) können privilegiert, vertraulich oder
nach dem Gesetz geheime Informationen enthalten. Sie sind ausschließlich
für den genannten Empfänger bestimmt, an den sie gerichtet sind. Falls
Sie nicht der genannte Empfänger oder einer seiner Angestellten oder
Mitarbeiter sind, oder diese Mitteilung irrtümlich erhalten haben,
bitten wir Sie, uns unverzüglich darüber in Kenntnis zu setzen und diese
Mitteilung sowie alle bestehenden Kopien dieser Mitteilung zu
vernichten. Wir weisen Sie darauf hin, dass das Lesen, Weiterleiten,
Verbreiten oder Kopieren dieser Mitteilung ohne die Zustimmung des
Absenders verboten ist.
Hi,
with the instances being on different machines this is not
a strict requirement.
However, it might be good practice to have a clear
one-to-one correlation of the onconfig file to the
IDS instance. One way of doing that is by naming
the onconfig file with the instance's server name - in which
case you would actually want to change the name of
the onconfig file. And then you'd have to change the
setting of the ONCONFIG environment variable as well ... :)
Regards,
Martin
--
Martin Fuerderer
IBM Informix Development Munich, Germany
Information Management
IBM Deutschland Entwicklung GmbH
Chairman of the Supervisory Board: Martin Jetter
Board of Management: Herbert Kircher
Corporate Seat: Boeblingen, Germany
Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294
ids-bounces@iiug.org wrote on 11.03.2008 09:58:41:
> Hi Martin,
>
> I think you have to change the name of the onconfig file, also.
>
> Regards
> Knuth
>
> On Tue, 2008-03-11 at 04:38 -0400, Martin Fuerderer wrote:
> > Hi,
> >
> > a) when changing the name of the server, also change the environment
> > variable INFORMIXSERVER accordingly. (There are 3 places: the
> > environment variable, the entry in onconfig file and the entry in
> > sqlhosts
> > file.)
> >
> > b) check the message log (see MSGPATH in your onconfig file).
> > It may contain useful information to determine the cause of the
> > problem.
> >
> > Regards,
> > Martin
> > --
> > Martin Fuerderer
> > IBM Informix Development Munich, Germany
> > Information Management
> >
> > IBM Deutschland Entwicklung GmbH
> > Chairman of the Supervisory Board: Martin Jetter
> > Board of Management: Herbert Kircher
> > Corporate Seat: Boeblingen, Germany
> > Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294
> >
> > ids-bounces@iiug.org wrote on 10.03.2008 23:11:22:
> >
> > > So much for it being an identical configuration as I was told by the
> > > sysadmin who prepared it:)
> > >
> > > The sqlexec entry was indeed missing from /etc/services.
> > >
> > > I added it and tried to rerun ontape -r and got this message (and
the
> > > prompt didn't return):
> > >
> > > oninit: Fatal error in shared memory creation
> > >
> > > Some one else stated to look at $INFORMIXDIR/etc/sqlhosts. On the
> > > system that the backup was created on is courtshm and I changed that
to
> > > courtbak on the system I am trying to restore it to. This matches
the
> > > name I changed servername to in onconfig to avoid the duplicate when
> > > restored.
> > >
> > > Thanks,
> > > Randy
> > >
> > > -----Original Message-----
> > > From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf
Of
> > > Art S. Kagel (Oninit)
> > > Sent: Monday, March 10, 2008 1:43 PM
> > > To: ids@iiug.org
> > > Subject: Re: Trying to restore a level 0 backup to a di.... [11560]
> > >
> > > Kennedy, Randy wrote:
> > > > Server: HP/UX 11.11
> > > > Informix: 9.4FC3
> > > >
> > > > We are attempting to 'test' our backup solution and received the
> > > > following error when doing an 'ontape -r' on the backup server.
> > > > Physical restore failed - could not fork server connection,
SQLCODE
> > > > -931
> > > >
> > > > We have the tapedev configured as a file on a remote server
> > > > (/remote/recovery/informix/level0) on both servers.
> > > > The hardware and chunk links are identical on both servers.
> > > > The onconfig & host has different server names so when I bring up
the
> > > > backup (if I can restore the backup), they won't have same names.
> > > > It is a single instance with 24 dbspaces & 7 non system databases.
> > > > The backups have run without a problem for a couple of years and
have
> > > > never needed to test recovery (and didn't have the downtime to
really
> > > > try). Now we have this identical server and want to ensure our
backup
> > > > solution actually works. Still not sure:)
> > > >
> > > > I have read through the Backup & Restore guide several times for
> > > > Ontape restores and it doesn't say a whole lot, run ontape -r and
you
> > > > should be good to go.
> > > >
> > > > I pressed enter at the mount tape prompt, entered yes to continue
at
> > > > next prompt, no to backup logs on disk, it ran for about 30
minutes
> > > > then I noticed the error message. Below is the complete attempt.
Any
> > > > help would be greatly appreciated.
> > >
> > > This is the clue:
> > > > Physical restore failed - could not fork server connection,
SQLCODE
> > > > -931
> > > >
> > >
> > > Finderr -931 would report that the servicename for the server in the
> > > sqlhosts file is not found in /etc/services.
> > >
> > > Art S. Kagel
> > > Oninit
> > >
> > >
========================================================================
> >
> > > ===================
> > > Please access the attached hyperlink for an important electronic
> > > communications disclaimer:
> > >
> > > http://www.oninit.com/home/disclaimer.php
> > >
> > >
========================================================================
> >
> > > ===================
> > >
> > >
************************************************************************
> >
> > > *******
> > > Forum Note: Use "Reply" to post a response in the discussion forum.
> > >
> > > See you at the IIUG Informix 2008 Conference
> > > The Power Conference for Informix Professionals
> > > April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> > > http://www.iiug.org/conf
> > > Registration Now Open!!
> > >
> > >
> > >
> >
> >
>
*******************************************************************************
> > > Forum Note: Use "Reply" to post a response in the discussion forum.
> > >
> > > See you at the IIUG Informix 2008 Conference
> > > The Power Conference for Informix Professionals
> > > April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> > > http://www.iiug.org/conf
> > > Registration Now Open!!
> >
> >
> >
>
*******************************************************************************
> > Forum Note: Use "Reply" to post a response in the discussion forum.
> >
> > See you at the IIUG Informix 2008 Conference
> > The Power Conference for Informix Professionals
> > April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> > http://www.iiug.org/conf
> > Registration Now Open!!
>
> Knuth Hartlieb
> IMPIRE AG
> "AGROB-Medienpark" Geb. 5
> Münchener Straße 101B
> D-85737 Ismaning
>
> Phone: +49 (0) 89-4 52 25-4 61
> Fax: +49 (0) 89-4 52 25-2 90
> Mobil: +49 (0) 1 72-4 38 35 00
> E-mail: knuth.hartlieb@impire.de
> Skype: kjhartlieb
> ITToolbox: kjhartlieb
>
> Besuchen Sie uns unter
>
> www.impire.de & www.bundesliga-datenbank.de
>
> IMPIRE AG: Sitz: D-85737 Ismaning, Amtsgericht München HR B 152 835;
> Vorstand: Hartmut Braun (Vorsitzender), Harald Keilbach;
> Aufsichtsratsvorsitzender: Bernd Künzler
>
Ok, I will try to explain what I did and where I stand.
Both systems are identical hardware. I will call the production system that
the backup was taken from PROD and the new system that I am trying to restore
it to BAK.
On Prod, the settings are:
Servername - courtshm
Onconfig - onconfig.courtshm
Everything works just fine.
On BAK, the system was supposed to be a mirror of PROD (and it was for the
most part, /etc/services entry withstanding)
The chunks are raw devices referenced as links, the links are identical on
both systems.
I modified the onconfig file as follows (before attempting restore)
Changed the name to onconfig.courtsbak
changed servername to courtsbak
I modified env variables for informixserver and onconfig to reference
courtsbak & onconfig.courtsbak.
I modified sqlhosts to reference courtsbak name
I left the tapedev the same so it would point to the backup 'file' created by
ontape from PROD
Executed 'ontape -r' (a 2nd time after adding the sqlexec entry to
/etc/services)
Received the "oninit: Fatal error in shared memory creation" error as I posted
last night. Went home and left it.
This morning, the prompt returned so I did an 'onstat -' to see where I was.
It showed block checkpoint, so I shut down server, changed tapedev to
/dev/null and did the oninit and get the following:
=>oninit
oninit: Cannot open chunk '/db/links/llogdbs'. errno = 2
oninit: Cannot open chunk '/db/links/casedbs'. errno = 2
oninit: Cannot open chunk '/db/links/physdbs'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs1'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs3'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs4'. errno = 2
oninit: Cannot open chunk '/db/links/chrgdbs'. errno = 2
oninit: Cannot open chunk '/db/links/dwdbs'. errno = 2
oninit: Cannot open chunk '/db/links/clndrdbs'. errno = 2
oninit: Cannot open chunk '/db/links/ctntndbs'. errno = 2
oninit: Cannot open chunk '/db/links/eventdbs'. errno = 2
oninit: Cannot open chunk '/db/links/masterdbs'. errno = 2
oninit: Cannot open chunk '/db/links/peopledbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs3'. errno = 2
oninit: Cannot open chunk '/db/links/smat2dbs1'. errno = 2
oninit: Cannot open chunk '/db/links/awdbs'. errno = 2
oninit: Cannot open chunk '/db/links/rcptmstrdbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs1'. errno = 2
oninit: Cannot open chunk '/db/links/eventdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/sccdbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/testdbs'. errno = 2
Can Not Open Primary Chunk /db/links/llogdbs
oninit: Fatal error in shared memory initialization
Here is the log file:
Tue Mar 11 09:47:14 2008
09:47:14 Event alarms enabled. ALARMPROG = '/db/scripts/no_log.sh'
09:47:14 Booting Language <c> from module <>
09:47:14 Loading Module <CNULL>
09:47:14 Booting Language <builtin> from module <>
09:47:14 Loading Module <BUILTINNULL>
09:47:23 IBM Informix Dynamic Server Version 9.40.FC3 Software Serial Number
AAA#B000000
09:47:23 Assert Warning: chunk failed sanity check
09:47:23 IBM Informix Dynamic Server Version 9.40.FC3
09:47:23 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:23 Results: Chunk 2 is being taken OFFLINE.
09:47:23 Action: Restore chunk from archive.
09:47:23 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:23 See Also: /tmp/af.407b79a
09:47:24 chunk failed sanity check
09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
09:47:24 Assert Warning: chunk failed sanity check
09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:24 Results: Chunk 3 is being taken OFFLINE.
09:47:24 Action: Restore chunk from archive.
09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:24 See Also: /tmp/af.407b79a
09:47:25 chunk failed sanity check
09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
09:47:24 Assert Warning: chunk failed sanity check
09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:24 Results: Chunk 3 is being taken OFFLINE.
09:47:24 Action: Restore chunk from archive.
09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:24 See Also: /tmp/af.407b79a
09:47:25 chunk failed sanity check
09:47:25 I/O error, Primary Chunk '/db/links/casedbs' -- Offline (sanity)
09:47:25 Assert Warning: chunk failed sanity check
-----repeats for every chunk----------
09:47:47 mt_shm_remove: WARNING: may not have removed all/correct segments
What am I missing/doing wrong??
Thanks,
Randy
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Martin
Fuerderer
Sent: Tuesday, March 11, 2008 3:00 AM
To: ids@iiug.org
Subject: RE: Trying to restore a level 0 backup to a di.... [11572]
Hi,
with the instances being on different machines this is not a strict
requirement.
However, it might be good practice to have a clear one-to-one correlation of
the onconfig file to the IDS instance. One way of doing that is by naming the
onconfig file with the instance's server name - in which case you would
actually want to change the name of the onconfig file. And then you'd have to
change the setting of the ONCONFIG environment variable as well ... :)
Regards,
Martin
--
Martin Fuerderer
IBM Informix Development Munich, Germany Information Management
IBM Deutschland Entwicklung GmbH
Chairman of the Supervisory Board: Martin Jetter Board of Management: Herbert
Kircher Corporate Seat: Boeblingen, Germany
Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294
ids-bounces@iiug.org wrote on 11.03.2008 09:58:41:
> Hi Martin,
>
> I think you have to change the name of the onconfig file, also.
>
> Regards
> Knuth
>
> On Tue, 2008-03-11 at 04:38 -0400, Martin Fuerderer wrote:
> > Hi,
> >
> > a) when changing the name of the server, also change the environment
> > variable INFORMIXSERVER accordingly. (There are 3 places: the
> > environment variable, the entry in onconfig file and the entry in
> > sqlhosts
> > file.)
> >
> > b) check the message log (see MSGPATH in your onconfig file).
> > It may contain useful information to determine the cause of the
> > problem.
> >
> > Regards,
> > Martin
> > --
> > Martin Fuerderer
> > IBM Informix Development Munich, Germany Information Management
> >
> > IBM Deutschland Entwicklung GmbH
> > Chairman of the Supervisory Board: Martin Jetter Board of
> > Management: Herbert Kircher Corporate Seat: Boeblingen, Germany
> > Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294
> >
> > ids-bounces@iiug.org wrote
Hi,
a) check your links and underlying (raw) devices.
Access permissions? File (or device) ownership correct? ...
b) check that IDS is installed correctly, i.e. binaries
like oninit, etc. have the correct setting of ownership
and access rights.
c) check that as user informix you can read from and write
to the links (that you use as chunk names).
E.g. use the system utility "dd" to do that, like
"dd if=/dev/zero of=/db/links/casedbs bs=2048 count=500"
d) you may do the same read/write check for the underlying
devices as well, although if it works with the links, it should
with the devices as well.
When all this is correct and works, then you should no longer
see the system error 2 (which means "No such file or
directory").
You may have to repeat the restore (as I'm not sure whether that
has worked at all ... it doesn't look like).
Regards,
Martin
--
Martin Fuerderer
IBM Informix Development Munich, Germany
Information Management
Notes: Martin Fuerderer/Germany/IBM@IBMDE
IBM Deutschland Entwicklung GmbH
Chairman of the Supervisory Board: Martin Jetter
Board of Management: Herbert Kircher
Corporate Seat: Boeblingen, Germany
Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294
"Kennedy, Randy" <RKennedy@scottsdaleaz.gov>
Sent by: ids-bounces@iiug.org
11.03.2008 18:19
Please respond to
ids@iiug.org
To
ids@iiug.org
cc
Subject
RE: Trying to restore a level 0 backup to a di.... [11574]
Ok, I will try to explain what I did and where I stand.
Both systems are identical hardware. I will call the production system
that
the backup was taken from PROD and the new system that I am trying to
restore
it to BAK.
On Prod, the settings are:
Servername - courtshm
Onconfig - onconfig.courtshm
Everything works just fine.
On BAK, the system was supposed to be a mirror of PROD (and it was for the
most part, /etc/services entry withstanding)
The chunks are raw devices referenced as links, the links are identical on
both systems.
I modified the onconfig file as follows (before attempting restore)
Changed the name to onconfig.courtsbak
changed servername to courtsbak
I modified env variables for informixserver and onconfig to reference
courtsbak & onconfig.courtsbak.
I modified sqlhosts to reference courtsbak name
I left the tapedev the same so it would point to the backup 'file' created
by
ontape from PROD
Executed 'ontape -r' (a 2nd time after adding the sqlexec entry to
/etc/services)
Received the "oninit: Fatal error in shared memory creation" error as I
posted
last night. Went home and left it.
This morning, the prompt returned so I did an 'onstat -' to see where I
was.
It showed block checkpoint, so I shut down server, changed tapedev to
/dev/null and did the oninit and get the following:
=>oninit
oninit: Cannot open chunk '/db/links/llogdbs'. errno = 2
oninit: Cannot open chunk '/db/links/casedbs'. errno = 2
oninit: Cannot open chunk '/db/links/physdbs'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs1'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs3'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs4'. errno = 2
oninit: Cannot open chunk '/db/links/chrgdbs'. errno = 2
oninit: Cannot open chunk '/db/links/dwdbs'. errno = 2
oninit: Cannot open chunk '/db/links/clndrdbs'. errno = 2
oninit: Cannot open chunk '/db/links/ctntndbs'. errno = 2
oninit: Cannot open chunk '/db/links/eventdbs'. errno = 2
oninit: Cannot open chunk '/db/links/masterdbs'. errno = 2
oninit: Cannot open chunk '/db/links/peopledbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs3'. errno = 2
oninit: Cannot open chunk '/db/links/smat2dbs1'. errno = 2
oninit: Cannot open chunk '/db/links/awdbs'. errno = 2
oninit: Cannot open chunk '/db/links/rcptmstrdbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs1'. errno = 2
oninit: Cannot open chunk '/db/links/eventdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/sccdbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/testdbs'. errno = 2
Can Not Open Primary Chunk /db/links/llogdbs
oninit: Fatal error in shared memory initialization
Here is the log file:
Tue Mar 11 09:47:14 2008
09:47:14 Event alarms enabled. ALARMPROG = '/db/scripts/no_log.sh'
09:47:14 Booting Language <c> from module <>
09:47:14 Loading Module <CNULL>
09:47:14 Booting Language <builtin> from module <>
09:47:14 Loading Module <BUILTINNULL>
09:47:23 IBM Informix Dynamic Server Version 9.40.FC3 Software Serial
Number
AAA#B000000
09:47:23 Assert Warning: chunk failed sanity check
09:47:23 IBM Informix Dynamic Server Version 9.40.FC3
09:47:23 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:23 Results: Chunk 2 is being taken OFFLINE.
09:47:23 Action: Restore chunk from archive.
09:47:23 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:23 See Also: /tmp/af.407b79a
09:47:24 chunk failed sanity check
09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
09:47:24 Assert Warning: chunk failed sanity check
09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:24 Results: Chunk 3 is being taken OFFLINE.
09:47:24 Action: Restore chunk from archive.
09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:24 See Also: /tmp/af.407b79a
09:47:25 chunk failed sanity check
09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
09:47:24 Assert Warning: chunk failed sanity check
09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:24 Results: Chunk 3 is being taken OFFLINE.
09:47:24 Action: Restore chunk from archive.
09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:24 See Also: /tmp/af.407b79a
09:47:25 chunk failed sanity check
09:47:25 I/O error, Primary Chunk '/db/links/casedbs' -- Offline (sanity)
09:47:25 Assert Warning: chunk failed sanity check
-----repeats for every chunk----------
09:47:47 mt_shm_remove: WARNING: may not have removed all/correct segments
What am I missing/doing wrong??
Thanks,
Randy
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
Martin
Fuerderer
Sent: Tuesday, March 11, 2008 3:00 AM
To: ids@iiug.org
Subject: RE: Trying to restore a level 0 backup to a di.... [11572]
Hi,
with the instances being on different machines this is not a strict
requirement.
However, it might be good practice to have a clear one-to-one correlation
of
the onconfig file to the IDS instance. One way of doing that is by naming
the@
Kennedy, Randy wrote:
Errno = 2 is a permissions error. Make sure the chunks that the links
point to are owned by informix and group informix and that their
permissions are 0666. Also the links should also be owned by informix
and group informix with privs 777.
Art S. Kagel
Oninit
> Ok, I will try to explain what I did and where I stand.
>
> Both systems are identical hardware. I will call the production system that
> the backup was taken from PROD and the new system that I am trying to restore
> it to BAK.
>
> On Prod, the settings are:
> Servername - courtshm
> Onconfig - onconfig.courtshm
> Everything works just fine.
>
> On BAK, the system was supposed to be a mirror of PROD (and it was for the
> most part, /etc/services entry withstanding)
> The chunks are raw devices referenced as links, the links are identical on
> both systems.
> I modified the onconfig file as follows (before attempting restore)
> Changed the name to onconfig.courtsbak
> changed servername to courtsbak
> I modified env variables for informixserver and onconfig to reference
> courtsbak & onconfig.courtsbak.
> I modified sqlhosts to reference courtsbak name
> I left the tapedev the same so it would point to the backup 'file' created by
> ontape from PROD
> Executed 'ontape -r' (a 2nd time after adding the sqlexec entry to
> /etc/services)
> Received the "oninit: Fatal error in shared memory creation" error as I
posted
> last night. Went home and left it.
> This morning, the prompt returned so I did an 'onstat -' to see where I was.
> It showed block checkpoint, so I shut down server, changed tapedev to
> /dev/null and did the oninit and get the following:
> =>oninit
> oninit: Cannot open chunk '/db/links/llogdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/casedbs'. errno = 2
> oninit: Cannot open chunk '/db/links/physdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/tempdbs1'. errno = 2
> oninit: Cannot open chunk '/db/links/tempdbs2'. errno = 2
> oninit: Cannot open chunk '/db/links/tempdbs3'. errno = 2
> oninit: Cannot open chunk '/db/links/tempdbs4'. errno = 2
> oninit: Cannot open chunk '/db/links/chrgdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/dwdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/clndrdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/ctntndbs'. errno = 2
> oninit: Cannot open chunk '/db/links/eventdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/masterdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/peopledbs'. errno = 2
> oninit: Cannot open chunk '/db/links/1gbdbs3'. errno = 2
> oninit: Cannot open chunk '/db/links/smat2dbs1'. errno = 2
> oninit: Cannot open chunk '/db/links/awdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/rcptmstrdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/1gbdbs1'. errno = 2
> oninit: Cannot open chunk '/db/links/eventdbs2'. errno = 2
> oninit: Cannot open chunk '/db/links/sccdbs'. errno = 2
> oninit: Cannot open chunk '/db/links/1gbdbs2'. errno = 2
> oninit: Cannot open chunk '/db/links/testdbs'. errno = 2
> Can Not Open Primary Chunk /db/links/llogdbs
> oninit: Fatal error in shared memory initialization
>
> Here is the log file:
> Tue Mar 11 09:47:14 2008
>
> 09:47:14 Event alarms enabled. ALARMPROG = '/db/scripts/no_log.sh'
> 09:47:14 Booting Language <c> from module <>
> 09:47:14 Loading Module <CNULL>
> 09:47:14 Booting Language <builtin> from module <>
> 09:47:14 Loading Module <BUILTINNULL>
> 09:47:23 IBM Informix Dynamic Server Version 9.40.FC3 Software Serial Number
> AAA#B000000
> 09:47:23 Assert Warning: chunk failed sanity check
>
> 09:47:23 IBM Informix Dynamic Server Version 9.40.FC3
> 09:47:23 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
>
> Thread(31, main_loop(), c000000015743028, 1)
>
> File: rspartn.c Line: 7733
> 09:47:23 Results: Chunk 2 is being taken OFFLINE.
> 09:47:23 Action: Restore chunk from archive.
> 09:47:23 stack trace for pid 29728 written to /tmp/af.407b79a
> 09:47:23 See Also: /tmp/af.407b79a
> 09:47:24 chunk failed sanity check
>
> 09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
> 09:47:24 Assert Warning: chunk failed sanity check
>
> 09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
> 09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
>
> Thread(31, main_loop(), c000000015743028, 1)
>
> File: rspartn.c Line: 7733
> 09:47:24 Results: Chunk 3 is being taken OFFLINE.
> 09:47:24 Action: Restore chunk from archive.
> 09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
> 09:47:24 See Also: /tmp/af.407b79a
> 09:47:25 chunk failed sanity check
>
> 09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
> 09:47:24 Assert Warning: chunk failed sanity check
>
> 09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
> 09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
>
> Thread(31, main_loop(), c000000015743028, 1)
>
> File: rspartn.c Line: 7733
> 09:47:24 Results: Chunk 3 is being taken OFFLINE.
> 09:47:24 Action: Restore chunk from archive.
> 09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
> 09:47:24 See Also: /tmp/af.407b79a
> 09:47:25 chunk failed sanity check
>
> 09:47:25 I/O error, Primary Chunk '/db/links/casedbs' -- Offline (sanity)
> 09:47:25 Assert Warning: chunk failed sanity check
> -----repeats for every chunk----------
>
> 09:47:47 mt_shm_remove: WARNING: may not have removed all/correct segments
>
> What am I missing/doing wrong??
>
> Thanks,
> Randy
>
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Martin
> Fuerderer
> Sent: Tuesday, March 11, 2008 3:00 AM
> To: ids@iiug.org
> Subject: RE: Trying to restore a level 0 backup to a di.... [11572]
>
> Hi,
>
> with the instances being on different machines this is not a strict
> requirement.
>
> However, it might be good practice to have a clear one-to-one correlation of
> the onconfig file to the IDS instance. One way of doing that is by naming the
> onconfig file with the instance's server name - in which case you would
> actually want to change the name of the onconfig file. And then you'd have to
> change the setting of the ONCONFIG environment variable as well ... :)
>
> Regards,
> Martin
>
> ------------------------------------------------------------------------
>
> See you at the IIUG Informix 2008 Conference
> The Power Conference for Informix Professionals
> April 27 - 30, 2008 Marriott Overland Park (Kansas City), Kansas
> http://www.iiug.org/conf
> Registration Now Open!!
================================================================================
===========
Please access the attached hyperlink for an important electronic
communications disclaimer:
http://www.oninit.com/home/disclaimer.php
================================================================================
===========
Check the permissions and ownership of the underlying raw devices. You may
have to wade thru multiple levels of links, but it is the raw devices you have
to modify. Thee permission must be rw-rw---- for each device and the owner and
group must be informix.
"Kennedy, Randy" <RKennedy@scottsdaleaz.gov> wrote:
Ok, I will try to explain what I did and where I stand.
Both systems are identical hardware. I will call the production system that
the backup was taken from PROD and the new system that I am trying to restore
it to BAK.
On Prod, the settings are:
Servername - courtshm
Onconfig - onconfig.courtshm
Everything works just fine.
On BAK, the system was supposed to be a mirror of PROD (and it was for the
most part, /etc/services entry withstanding)
The chunks are raw devices referenced as links, the links are identical on
both systems.
I modified the onconfig file as follows (before attempting restore)
Changed the name to onconfig.courtsbak
changed servername to courtsbak
I modified env variables for informixserver and onconfig to reference
courtsbak & onconfig.courtsbak.
I modified sqlhosts to reference courtsbak name
I left the tapedev the same so it would point to the backup 'file' created by
ontape from PROD
Executed 'ontape -r' (a 2nd time after adding the sqlexec entry to
/etc/services)
Received the "oninit: Fatal error in shared memory creation" error as I posted
last night. Went home and left it.
This morning, the prompt returned so I did an 'onstat -' to see where I was.
It showed block checkpoint, so I shut down server, changed tapedev to
/dev/null and did the oninit and get the following:
=>oninit
oninit: Cannot open chunk '/db/links/llogdbs'. errno = 2
oninit: Cannot open chunk '/db/links/casedbs'. errno = 2
oninit: Cannot open chunk '/db/links/physdbs'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs1'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs3'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs4'. errno = 2
oninit: Cannot open chunk '/db/links/chrgdbs'. errno = 2
oninit: Cannot open chunk '/db/links/dwdbs'. errno = 2
oninit: Cannot open chunk '/db/links/clndrdbs'. errno = 2
oninit: Cannot open chunk '/db/links/ctntndbs'. errno = 2
oninit: Cannot open chunk '/db/links/eventdbs'. errno = 2
oninit: Cannot open chunk '/db/links/masterdbs'. errno = 2
oninit: Cannot open chunk '/db/links/peopledbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs3'. errno = 2
oninit: Cannot open chunk '/db/links/smat2dbs1'. errno = 2
oninit: Cannot open chunk '/db/links/awdbs'. errno = 2
oninit: Cannot open chunk '/db/links/rcptmstrdbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs1'. errno = 2
oninit: Cannot open chunk '/db/links/eventdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/sccdbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/testdbs'. errno = 2
Can Not Open Primary Chunk /db/links/llogdbs
oninit: Fatal error in shared memory initialization
Here is the log file:
Tue Mar 11 09:47:14 2008
09:47:14 Event alarms enabled. ALARMPROG = '/db/scripts/no_log.sh'
09:47:14 Booting Language from module <>
09:47:14 Loading Module
09:47:14 Booting Language from module <>
09:47:14 Loading Module
09:47:23 IBM Informix Dynamic Server Version 9.40.FC3 Software Serial Number
AAA#B000000
09:47:23 Assert Warning: chunk failed sanity check
09:47:23 IBM Informix Dynamic Server Version 9.40.FC3
09:47:23 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:23 Results: Chunk 2 is being taken OFFLINE.
09:47:23 Action: Restore chunk from archive.
09:47:23 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:23 See Also: /tmp/af.407b79a
09:47:24 chunk failed sanity check
09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
09:47:24 Assert Warning: chunk failed sanity check
09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:24 Results: Chunk 3 is being taken OFFLINE.
09:47:24 Action: Restore chunk from archive.
09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:24 See Also: /tmp/af.407b79a
09:47:25 chunk failed sanity check
09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
09:47:24 Assert Warning: chunk failed sanity check
09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:24 Results: Chunk 3 is being taken OFFLINE.
09:47:24 Action: Restore chunk from archive.
09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:24 See Also: /tmp/af.407b79a
09:47:25 chunk failed sanity check
09:47:25 I/O error, Primary Chunk '/db/links/casedbs' -- Offline (sanity)
09:47:25 Assert Warning: chunk failed sanity check
-----repeats for every chunk----------
09:47:47 mt_shm_remove: WARNING: may not have removed all/correct segments
What am I missing/doing wrong??
Thanks,
Randy
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Martin
Fuerderer
Sent: Tuesday, March 11, 2008 3:00 AM
To: ids@iiug.org
Subject: RE: Trying to restore a level 0 backup to a di.... [11572]
Hi,
with the instances being on different machines this is not a strict
requirement.
However, it might be good practice to have a clear one-to-one correlation of
the onconfig file to the IDS instance. One way of doing that is by naming the
onconfig file with the instance's server name - in which case you would
actually want to change the name of the onconfig file. And then you'd have to
change the setting of the ONCONFIG environment variable as well ... :)
Regards,
Martin
--
Martin Fuerderer
IBM Informix Development Munich, Germany Information Management
IBM Deutschland Entwicklung GmbH
Chairman of the Supervisory Board: Martin Jetter Board of Management: Herbert
Kircher Corporate Seat: Boeblingen, Germany
Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294
ids-bounces@iiug.org wrote on 11.03.2008 09:58:41:
> Hi Martin,
>
> I think you have to change the name of the onconfig file, also.
>
> Regards
> Knuth
>
> On Tue, 2008-03-11 at 04:38 -0400, Martin Fuerderer wrote:
> > Hi,
> >
> > a) when changing the name of the server, also change the environment
> > variable INFORMIXSERVER accordingly. (There are 3 places: the
> > environment variable, the entry in onconfig file and the entry in
> > sqlhosts
> > file.)
> >
> > b) check the message log (see MSGPATH in your onconfig file).
> > It may contain useful information to determine the cause of the
> > problem.
> >
> > Regards,
> > Martin
> > --
> > Martin Fuerderer
> > IBM Informix Development Munich, Germany I
SUCCESS!!!!
Thanks to all who replied, I had several issues get me on this.
The missing /etc/services entry as pointed out earlier and then the underlying
raw devices of the links. Again the sysadmin told me he mirrored the
production box on the backup box and was about 99% correct. The devices were
all defined, the correct size, but didn't have correct ownership/read-write
attributes.
I now know that our backups are useful for restoration and have experience
doing it.
Next up, HDR to the backup box:)
Thanks again,
Randy
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Walter
Lowich
Sent: Tuesday, March 11, 2008 12:33 PM
To: ids@iiug.org
Subject: RE: Trying to restore a level 0 backup to a di.... [11577]
Check the permissions and ownership of the underlying raw devices. You may
have to wade thru multiple levels of links, but it is the raw devices you have
to modify. Thee permission must be rw-rw---- for each device and the owner and
group must be informix.
"Kennedy, Randy" <RKennedy@scottsdaleaz.gov> wrote:
Ok, I will try to explain what I did and where I stand.
Both systems are identical hardware. I will call the production system that
the backup was taken from PROD and the new system that I am trying to restore
it to BAK.
On Prod, the settings are:
Servername - courtshm
Onconfig - onconfig.courtshm
Everything works just fine.
On BAK, the system was supposed to be a mirror of PROD (and it was for the
most part, /etc/services entry withstanding) The chunks are raw devices
referenced as links, the links are identical on both systems.
I modified the onconfig file as follows (before attempting restore) Changed
the name to onconfig.courtsbak changed servername to courtsbak I modified env
variables for informixserver and onconfig to reference courtsbak &
onconfig.courtsbak.
I modified sqlhosts to reference courtsbak name I left the tapedev the same so
it would point to the backup 'file' created by ontape from PROD Executed
'ontape -r' (a 2nd time after adding the sqlexec entry to
/etc/services)
Received the "oninit: Fatal error in shared memory creation" error as I posted
last night. Went home and left it.
This morning, the prompt returned so I did an 'onstat -' to see where I was.
It showed block checkpoint, so I shut down server, changed tapedev to
/dev/null and did the oninit and get the following:
=>oninit
oninit: Cannot open chunk '/db/links/llogdbs'. errno = 2
oninit: Cannot open chunk '/db/links/casedbs'. errno = 2
oninit: Cannot open chunk '/db/links/physdbs'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs1'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs3'. errno = 2
oninit: Cannot open chunk '/db/links/tempdbs4'. errno = 2
oninit: Cannot open chunk '/db/links/chrgdbs'. errno = 2
oninit: Cannot open chunk '/db/links/dwdbs'. errno = 2
oninit: Cannot open chunk '/db/links/clndrdbs'. errno = 2
oninit: Cannot open chunk '/db/links/ctntndbs'. errno = 2
oninit: Cannot open chunk '/db/links/eventdbs'. errno = 2
oninit: Cannot open chunk '/db/links/masterdbs'. errno = 2
oninit: Cannot open chunk '/db/links/peopledbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs3'. errno = 2
oninit: Cannot open chunk '/db/links/smat2dbs1'. errno = 2
oninit: Cannot open chunk '/db/links/awdbs'. errno = 2
oninit: Cannot open chunk '/db/links/rcptmstrdbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs1'. errno = 2
oninit: Cannot open chunk '/db/links/eventdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/sccdbs'. errno = 2
oninit: Cannot open chunk '/db/links/1gbdbs2'. errno = 2
oninit: Cannot open chunk '/db/links/testdbs'. errno = 2 Can Not Open Primary
Chunk /db/links/llogdbs
oninit: Fatal error in shared memory initialization
Here is the log file:
Tue Mar 11 09:47:14 2008
09:47:14 Event alarms enabled. ALARMPROG = '/db/scripts/no_log.sh'
09:47:14 Booting Language from module <>
09:47:14 Loading Module
09:47:14 Booting Language from module <>
09:47:14 Loading Module
09:47:23 IBM Informix Dynamic Server Version 9.40.FC3 Software Serial Number
AAA#B000000
09:47:23 Assert Warning: chunk failed sanity check
09:47:23 IBM Informix Dynamic Server Version 9.40.FC3
09:47:23 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:23 Results: Chunk 2 is being taken OFFLINE.
09:47:23 Action: Restore chunk from archive.
09:47:23 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:23 See Also: /tmp/af.407b79a
09:47:24 chunk failed sanity check
09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
09:47:24 Assert Warning: chunk failed sanity check
09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:24 Results: Chunk 3 is being taken OFFLINE.
09:47:24 Action: Restore chunk from archive.
09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:24 See Also: /tmp/af.407b79a
09:47:25 chunk failed sanity check
09:47:24 I/O error, Primary Chunk '/db/links/llogdbs' -- Offline (sanity)
09:47:24 Assert Warning: chunk failed sanity check
09:47:24 IBM Informix Dynamic Server Version 9.40.FC3
09:47:24 Who: Session(1, informix@hp9ux2, 0, c000000015785028)
Thread(31, main_loop(), c000000015743028, 1)
File: rspartn.c Line: 7733
09:47:24 Results: Chunk 3 is being taken OFFLINE.
09:47:24 Action: Restore chunk from archive.
09:47:24 stack trace for pid 29728 written to /tmp/af.407b79a
09:47:24 See Also: /tmp/af.407b79a
09:47:25 chunk failed sanity check
09:47:25 I/O error, Primary Chunk '/db/links/casedbs' -- Offline (sanity)
09:47:25 Assert Warning: chunk failed sanity check -----repeats for every
chunk----------
09:47:47 mt_shm_remove: WARNING: may not have removed all/correct segments
What am I missing/doing wrong??
Thanks,
Randy
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Martin
Fuerderer
Sent: Tuesday, March 11, 2008 3:00 AM
To: ids@iiug.org
Subject: RE: Trying to restore a level 0 backup to a di.... [11572]
Hi,
with the instances being on different machines this is not a strict
requirement.
However, it might be good practice to have a clear one-to-one correlation of
the onconfig file to the IDS instance. One way of doing that is by naming the
onconfig file with the instance's server name - in which case you would
actually want to change the name of the onconfig file. And then you'd have to
change the setting of the ONCONFIG environment variable as well ... :)
Regards,
Martin
--
Martin Fuerderer
IBM Informix Development Munich, Germany Information Management
IBM Deutschland Entwicklung GmbH
Chairman of the Supervisory Board: Martin Jetter Board of Management: Herbert
Kircher Corporate Seat: Boeblingen, Germany
Reg.-Gericht: Amtsgerich