Backup worked...restore doesn't...
Posted in 2007
CJ took an ontape -s -L 0 level-0 backup of server prd01 but restoring it to a second machine (sci01) failed with "Physical restore failed - could not fork server connection, SQLCODE -25588". Replies pointed at networking/connectivity config: check sqlhosts against /etc/services, /etc/hosts and /etc/hosts.equiv, make sure DBSERVERNAME/aliases in the copied onconfig were changed, verify chunk links/sizes, and (Art Kagel) avoid reusing the same service name for both the onipcshm and onsoctcp entries. The real cause turned out to be that IT had changed the server's IP address without telling him; after correcting /etc/hosts the restore succeeded.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Topics: Backup & Restore, Installation, Setup & Upgrades, Error Codes & Troubleshooting, Server Administration
Hi All,
I was able to get my backup done the other day just using ontape -s -L 0,
everything seemed just fine, total space of backup 497 GB. Now today I'm
trying to restore that same tape onto another server. Informix has been
installed, onconfig file was mirrored from original server, permissions all
look good, but ontape -r provides this error after normal prompts to continue
and backup logs. (I answered Y and N).
Physical restore failed - could not fork server connection, SQLCODE -25588
Any ideas?
Thanks,
CJ
check the sqlhosts ... making sure that the host name is correct ...
"CJ BUTCHER" <creighton_butcher@adaptec.com>
Sent by: ids-bounces@iiug.org
07/20/2007 03:49 PM
Please respond to
ids@iiug.org
To
ids@iiug.org
cc
Subject
Backup worked...restore doesn't... [9602]
Hi All,
I was able to get my backup done the other day just using ontape -s -L 0,
everything seemed just fine, total space of backup 497 GB. Now today I'm
trying to restore that same tape onto another server. Informix has been
installed, onconfig file was mirrored from original server, permissions
all
look good, but ontape -r provides this error after normal prompts to
continue
and backup logs. (I answered Y and N).
Physical restore failed - could not fork server connection, SQLCODE -25588
Any ideas?
Thanks,
CJ
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
Check the relationships between sqlhosts and /etc/services and /etc/hosts.
"CJ BUTCHER" <creighton_butcher@adaptec.com>
Sent by: ids-bounces@iiug.org
07/20/2007 02:49 PM
Please respond to
ids@iiug.org
To
ids@iiug.org
cc
Subject
Backup worked...restore doesn't... [9602]
Hi All,
I was able to get my backup done the other day just using ontape -s -L 0,
everything seemed just fine, total space of backup 497 GB. Now today I'm
trying to restore that same tape onto another server. Informix has been
installed, onconfig file was mirrored from original server, permissions
all
look good, but ontape -r provides this error after normal prompts to
continue
and backup logs. (I answered Y and N).
Physical restore failed - could not fork server connection, SQLCODE -25588
Any ideas?
Thanks,
CJ
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
I believe 25xxx is a network error.
/etc/services has to match $INFORMIXSQLHOSTS has to match
$INFORMIXSERVER has to match /etc/hosts has to match /etc/hosts.equiv
Bob Roussey
Unix / Informix Administration
Spirit Airlines
Robert.Roussey@SpiritAir.com
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of CJ
BUTCHER
Sent: Friday, July 20, 2007 3:49 PM
To: ids@iiug.org
Subject: Backup worked...restore doesn't... [9602]
Hi All,
I was able to get my backup done the other day just using ontape -s -L
0,
everything seemed just fine, total space of backup 497 GB. Now today I'm
trying to restore that same tape onto another server. Informix has been
installed, onconfig file was mirrored from original server, permissions
all
look good, but ontape -r provides this error after normal prompts to
continue
and backup logs. (I answered Y and N).
Physical restore failed - could not fork server connection, SQLCODE
-25588
Any ideas?
Thanks,
CJ
************************************************************************
*******
Forum Note: Use "Reply" to post a response in the discussion forum.
Thanks all for the ton or responses. I missed a point in my initial post:
The restore is on a hot backup server.
I backed up prd01
Need to restore to sci01
VALUES from sci01:
/etc/hosts
10.25.25.26 sci01 localhost.localdomain localhost
$INFORMIXSQLHOSTS
demo_on onipcshm on_hostname on_servernamedemo_se seipcpip se_hostname sqlexec
sci01scishm onipcshm sci01 sapinfSCI
sci01scitcp onsoctcp sci01 sapinfSCIprd01pritcp onsoctcp prd01 sapinfPRI s=6
sci01sectcp onsoctcp sci01 sapinfSEC s=6
prd01prdtcp onsoctcp prd01 sapinfPRD
$INFORMIXSERVER
sci01scishm
/etc/hosts.equiv
aimlss05
aimlss05.adaptec.com
192.168.83.174
VALUES from prd01:
/etc/hosts
127.0.0.1 localhost.localdomain localhost
10.25.16.26 prd01 prd01.adaptec.com
$INFORMIXSQLHOSTS
demo_on onipcshm on_hostname on_servernamedemo_se seipcpip se_hostname sqlexec
prd01prdshm onipcshm prd01 sapinfPRD
prd01prdtcp onsoctcp prd01 sapinfPRDprd01pritcp onsoctcp prd01 sapinfPRI s=6
sci01sectcp onsoctcp sci01 sapinfSEC s=6
$INFORMIXSERVER
prd01prdshm
/etc/hosts.equiv
(empty)
Is sapinfSCI in /etc/services on sci01?
"CJ BUTCHER" <creighton_butcher@adaptec.com>
Sent by: ids-bounces@iiug.org
07/20/2007 03:39 PM
Please respond to
ids@iiug.org
To
ids@iiug.org
cc
Subject
Re: RE: Backup worked...restore doesn't... [9606]
Thanks all for the ton or responses. I missed a point in my initial post:
The restore is on a hot backup server.
I backed up prd01
Need to restore to sci01
VALUES from sci01:
/etc/hosts
10.25.25.26 sci01 localhost.localdomain localhost
$INFORMIXSQLHOSTS
demo_on onipcshm on_hostname on_servernamedemo_se seipcpip se_hostname sqlexec
sci01scishm onipcshm sci01 sapinfSCI
sci01scitcp onsoctcp sci01 sapinfSCIprd01pritcp onsoctcp prd01 sapinfPRI s=6
sci01sectcp onsoctcp sci01 sapinfSEC s=6
prd01prdtcp onsoctcp prd01 sapinfPRD
$INFORMIXSERVER
sci01scishm
/etc/hosts.equiv
aimlss05
aimlss05.adaptec.com
192.168.83.174
VALUES from prd01:
/etc/hosts
127.0.0.1 localhost.localdomain localhost
10.25.16.26 prd01 prd01.adaptec.com
$INFORMIXSQLHOSTS
demo_on onipcshm on_hostname on_servernamedemo_se seipcpip se_hostname sqlexec
prd01prdshm onipcshm prd01 sapinfPRD
prd01prdtcp onsoctcp prd01 sapinfPRDprd01pritcp onsoctcp prd01 sapinfPRI s=6
sci01sectcp onsoctcp sci01 sapinfSEC s=6
$INFORMIXSERVER
prd01prdshm
/etc/hosts.equiv
(empty)
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
\
Hi Joseph, No, sapinfSCI is only on sci01, the production servers value is sapinfPRD, which is on both servers. CJ
ids-bounces@iiug.org wrote on 07/20/2007 10:39:11 PM: > Thanks all for the ton or responses. I missed a point in my initial post: > > The restore is on a hot backup server. > Probably a silly question but you did change the value of DBSERVERNAME and ALIAS in the $ONCONFIG after copying it over from PRD? HTH Tilman
FYI, don't do this:
sci01scishm onipcshm sci01 sapinfSCI
sci01scitcp onsoctcp sci01 sapinfSCI
Although the port/service for a shared memory connection is a dummy entry, I
have experienced conditions where either the servers or the clients get
confused when the dummy is the same as an actual port/service used by another
alias for the same instance. Make it anything but sapinfSCI, like
sapinfSCI_dum. That's likely not what's causing this problem, but you never
know 'til you fix it.
Art S. Kagel
----- Original Message -----
From: Cj Butcher <ids@iiug.org>
At: 7/20 16:39:34
Thanks all for the ton or responses. I missed a point in my initial post:
The restore is on a hot backup server.
I backed up prd01
Need to restore to sci01
VALUES from sci01:
/etc/hosts
10.25.25.26 sci01 localhost.localdomain localhost
$INFORMIXSQLHOSTS
demo_on onipcshm on_hostname on_servernamedemo_se seipcpip se_hostname sqlexec
sci01scishm onipcshm sci01 sapinfSCI
sci01scitcp onsoctcp sci01 sapinfSCIprd01pritcp onsoctcp prd01 sapinfPRI s=6
sci01sectcp onsoctcp sci01 sapinfSEC s=6
prd01prdtcp onsoctcp prd01 sapinfPRD
$INFORMIXSERVER
sci01scishm
/etc/hosts.equiv
aimlss05
aimlss05.adaptec.com
192.168.83.174
VALUES from prd01:
/etc/hosts
127.0.0.1 localhost.localdomain localhost
10.25.16.26 prd01 prd01.adaptec.com
$INFORMIXSQLHOSTS
demo_on onipcshm on_hostname on_servernamedemo_se seipcpip se_hostname sqlexec
prd01prdshm onipcshm prd01 sapinfPRD
prd01prdtcp onsoctcp prd01 sapinfPRDprd01pritcp onsoctcp prd01 sapinfPRI s=6
sci01sectcp onsoctcp sci01 sapinfSEC s=6
$INFORMIXSERVER
prd01prdshm
/etc/hosts.equiv
(empty)
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
Do all of the chunk links exist and are physically at least as large as the
originals on the source machine?
Art S. Kagel
----- Original Message -----
From: Cj Butcher <ids@iiug.org>
At: 7/20 15:49:41
Hi All,
I was able to get my backup done the other day just using ontape -s -L 0,
everything seemed just fine, total space of backup 497 GB. Now today I'm
trying to restore that same tape onto another server. Informix has been
installed, onconfig file was mirrored from original server, permissions all
look good, but ontape -r provides this error after normal prompts to continue
and backup logs. (I answered Y and N).
Physical restore failed - could not fork server connection, SQLCODE -25588
Any ideas?
Thanks,
CJ
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
Thanks for the suggestions Art, had some other priorities come up and now I'm
back to this one. The target server was up and running prior to it's physical
more and replication was pushing data from the source to the target so we feel
pretty good that all the chunks are there.
In this thread you suggested not doing the item below. When I remmed one out,
I got errors stating that sci01scishm wasn't in my SQLHost file and vise-versa
when I remmed the other one out:
sci01:informix 163> vi $INFORMIXSQLHOSTS
sci01:informix 164> onstat -d
shared memory not initialized for INFORMIXSERVER 'sci01scishm'
sci01:informix 165> oninit
oninit: DBSERVERNAME 'sci01scishm' not in sqlhosts file or sqlhosts file
contents
are damaged.
sci01:informix 166> vi $INFORMIXSQLHOSTS
sci01:informix 167> oninit
oninit: DBSERVERNAME 'sci01scitcp' not in sqlhosts file or sqlhosts file
contents
are damaged.
sci01:informix 168> vi $INFORMIXSQLHOSTS
sci01:informix 169> oninit
sci01:informix 170> onstat -d
shared memory not initialized for INFORMIXSERVER 'sci01scishm'
Current SQL Host file is:
demo_on onipcshm on_hostname on_servernamedemo_se seipcpip se_hostname sqlexec
sci01scishm onipcshm sci01 sapinfSCI
sci01scitcp onsoctcp sci01 sapinfSCIaimprd01pritcp onsoctcp aimprd01 sapinfPRI s=6
sci01sectcp onsoctcp sci01 sapinfSEC s=6
aimprd01prdtcp onsoctcp aimprd01 sapinfPRD
I'm confused on why the below settings are incorrect...any thoughts?
My onconfig file has the correct server names in it.
Thanks,
CJ
My suggestion is just to not use 'sapinfSCI' for the service name in the
onipcshm connection, but use some dummy name like 'dumshmport' that's not even
in /etc/services and only use 'sapinfSCI' for the TCP connection. I've seen
problems with IDS getting confused and returning spurious errors if the ipcshm
connection shares the port of an existing tlitcp or soctcp connection.
Art S. Kagel
----- Original Message -----
From: Cj Butcher <ids@iiug.org>
At: 7/26 13:27:04
Thanks for the suggestions Art, had some other priorities come up and now I'm
back to this one. The target server was up and running prior to it's physical
more and replication was pushing data from the source to the target so we feel
pretty good that all the chunks are there.
In this thread you suggested not doing the item below. When I remmed one out,
I got errors stating that sci01scishm wasn't in my SQLHost file and vise-versa
when I remmed the other one out:
sci01:informix 163> vi $INFORMIXSQLHOSTS
sci01:informix 164> onstat -d
shared memory not initialized for INFORMIXSERVER 'sci01scishm'
sci01:informix 165> oninit
oninit: DBSERVERNAME 'sci01scishm' not in sqlhosts file or sqlhosts file
contents
are damaged.
sci01:informix 166> vi $INFORMIXSQLHOSTS
sci01:informix 167> oninit
oninit: DBSERVERNAME 'sci01scitcp' not in sqlhosts file or sqlhosts file
contents
are damaged.
sci01:informix 168> vi $INFORMIXSQLHOSTS
sci01:informix 169> oninit
sci01:informix 170> onstat -d
shared memory not initialized for INFORMIXSERVER 'sci01scishm'
Current SQL Host file is:
demo_on onipcshm on_hostname on_servernamedemo_se seipcpip se_hostname sqlexec
sci01scishm onipcshm sci01 sapinfSCI
sci01scitcp onsoctcp sci01 sapinfSCIaimprd01pritcp onsoctcp aimprd01 sapinfPRI s=6
sci01sectcp onsoctcp sci01 sapinfSEC s=6
aimprd01prdtcp onsoctcp aimprd01 sapinfPRD
I'm confused on why the below settings are incorrect...any thoughts?
My onconfig file has the correct server names in it.
Thanks,
CJ
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
Turns out when the IT guys moved my server he changed the IP and didn't bother to let me know. Once I figure that out, I updated the hosts file with the new IP and the restore worked. Thank you all for you suggestions and help!
Related threads
- what's the matter with error -25588
- Re: "Network down" on fresh install
- The appl process cannot connect
- Error 25588 : IDS 7.31UD1 on Solaris8