onint - cannot open chunk error 2
On HP-UX Itanium with IDS 10.00.FC5, a fresh install ran oninit -i successfully but a subsequent oninit failed with 'chunk failed sanity check' (sane_chopen: pg_addr 0:0 != expected 1:2) and 'Cannot open chunk /opt/informix/dev/root.1, errno = 2'. Suggestions: check chunk ownership/permissions (660 informix:informix - already correct), and dd/od the raw device, which showed the root chunk's reserved pages were all zeros, pointing to something (possibly the SAN or a missing ROOTOFFSET to protect the device VTOC) overwriting the chunk. Advice was to re-init, retest with dd, or temporarily use a cooked file for rootdbs. A later dbimport also failed to connect. No resolution is recorded in the thread.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
IDS 10.00 FC5 on HPUX Itanium BL870c server migrating from IDS 10.00 HC5 on
HPUX PS-RISK rp7410 server
I'm setting IDS from scratch for the first time in 5+ years and my foggy
memory must be getting to me.
I'm in the process of migrating our main Admin App from a HPUX PA-RISK (IDS
10.00 HC5) server to a HPUX Itanium (IDS 10.00 FC5) server and because of a
number of changes in the platforms I'm setting up IDS from scratch and will be
dbexporting and dbimporting to move the data.
After installing IDS 10.00.FC5 on the Itanium server and configuring the
onconf file and setting up the chucks I tried oninit -i and it finished
without errors but when I try oninit I get errors.
What have I forgotten to do? Or what am I doing wrong?
onint
/var/adm/crash/af.405c0cd
09:12:30 sane_chopen:pg_addr 0:0 != expected 1:2
page
page
c00000002a3ef400: 00000000 00000000 00000000 00000000 ........ ........
...
09:12:30 Assert Warning: chunk failed sanity check
09:12:30 Who: Session(1, informix@zoom, 0, c000000028071030)
Thread(29, main_loop(), c000000028036030, 4)
File: rspartn.c Line: 8673
09:12:30 Results: Chunk 1 is being taken OFFLINE.
09:12:30 Action: Restore chunk from archive.
09:12:30 Stack for thread: 29 main_loop()
...
oninit: Cannot open chunk '/opt/informix/dev/root.1'. errno = 2
oninit: Cannot open chunk '/opt/informix/dev/root.1-m'. errno = 2
oninit: Fatal error in shared memory initialization
The chunks are on a SAN and I use the lvcreate & mknode to create the chunks.
The OS seems to be able to get to the chunk.
# dd if=/opt/informix/dev/root.1 of=/dev/null count=100 bs=1024
100+0 records in
100+0 records out
Any help would be appreciated.
John
Errno = 2 is a permissions problem. The privileges on the chunks MUST be
660 owner "informix" group "informix".
Art
On Thu, Oct 30, 2008 at 11:16 AM, John Adamski <adamski@graceland.edu>wrote:
> IDS 10.00 FC5 on HPUX Itanium BL870c server migrating from IDS 10.00 HC5 on
> HPUX PS-RISK rp7410 server
>
> I'm setting IDS from scratch for the first time in 5+ years and my foggy
> memory must be getting to me.
>
> I'm in the process of migrating our main Admin App from a HPUX PA-RISK (IDS
> 10.00 HC5) server to a HPUX Itanium (IDS 10.00 FC5) server and because of a
> number of changes in the platforms I'm setting up IDS from scratch and will
> be
> dbexporting and dbimporting to move the data.
>
> After installing IDS 10.00.FC5 on the Itanium server and configuring the
> onconf file and setting up the chucks I tried oninit -i and it finished
> without errors but when I try oninit I get errors.
>
> What have I forgotten to do? Or what am I doing wrong?
>
> onint
> /var/adm/crash/af.405c0cd
> 09:12:30 sane_chopen:pg_addr 0:0 != expected 1:2
> page
> page
> c00000002a3ef400: 00000000 00000000 00000000 00000000 ........ ........
> ....
> 09:12:30 Assert Warning: chunk failed sanity check
>
> 09:12:30 Who: Session(1, informix@zoom, 0, c000000028071030)
>
> Thread(29, main_loop(), c000000028036030, 4)
>
> File: rspartn.c Line: 8673
> 09:12:30 Results: Chunk 1 is being taken OFFLINE.
> 09:12:30 Action: Restore chunk from archive.
> 09:12:30 Stack for thread: 29 main_loop()
> ....
> oninit: Cannot open chunk '/opt/informix/dev/root.1'. errno = 2
> oninit: Cannot open chunk '/opt/informix/dev/root.1-m'. errno = 2
> oninit: Fatal error in shared memory initialization
>
> The chunks are on a SAN and I use the lvcreate & mknode to create the
> chunks.
> The OS seems to be able to get to the chunk.
>
> # dd if=/opt/informix/dev/root.1 of=/dev/null count=100 bs=1024
> 100+0 records in
> 100+0 records out
>
> Any help would be appreciated.
>
> John
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Art S. Kagel
Oninit (www.oninit.com)
IIUG Board of Directors (art@iiug.org)
Disclaimer: Please keep in mind that my own opinions are my own opinions and
do not reflect on my employer, Oninit, the IIUG, nor any other organization
with which I am associated either explicitly or implicitly. Neither do
those opinions reflect those of other individuals affiliated with any entity
with which I am affiliated nor those of the entities themselves.
Already checked that:
crw-rw---- 1 informix informix 64 0x020001 Sep 17 14:52 root.1
crw-rw---- 1 informix informix 64 0x020002 Sep 17 14:52 root.1-m
hmm maybe higher up the directory tree not right. I will compare with my old
server.
John
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Art Kagel
Sent: Thursday, October 30, 2008 10:45 AM
To: ids@iiug.org
Subject: Re: onint - cannot open chunk error 2 [13848]
Errno = 2 is a permissions problem. The privileges on the chunks MUST be
660 owner "informix" group "informix".
Art
On Thu, Oct 30, 2008 at 11:16 AM, John Adamski <adamski@graceland.edu>wrote:
> IDS 10.00 FC5 on HPUX Itanium BL870c server migrating from IDS 10.00 HC5 on
> HPUX PS-RISK rp7410 server
>
> I'm setting IDS from scratch for the first time in 5+ years and my foggy
> memory must be getting to me.
>
> I'm in the process of migrating our main Admin App from a HPUX PA-RISK (IDS
> 10.00 HC5) server to a HPUX Itanium (IDS 10.00 FC5) server and because of a
> number of changes in the platforms I'm setting up IDS from scratch and will
> be
> dbexporting and dbimporting to move the data.
>
> After installing IDS 10.00.FC5 on the Itanium server and configuring the
> onconf file and setting up the chucks I tried oninit -i and it finished
> without errors but when I try oninit I get errors.
>
> What have I forgotten to do? Or what am I doing wrong?
>
> onint
> /var/adm/crash/af.405c0cd
> 09:12:30 sane_chopen:pg_addr 0:0 != expected 1:2
> page
> page
> c00000002a3ef400: 00000000 00000000 00000000 00000000 ........ ........
> ....
> 09:12:30 Assert Warning: chunk failed sanity check
>
> 09:12:30 Who: Session(1, informix@zoom, 0, c000000028071030)
>
> Thread(29, main_loop(), c000000028036030, 4)
>
> File: rspartn.c Line: 8673
> 09:12:30 Results: Chunk 1 is being taken OFFLINE.
> 09:12:30 Action: Restore chunk from archive.
> 09:12:30 Stack for thread: 29 main_loop()
> ....
> oninit: Cannot open chunk '/opt/informix/dev/root.1'. errno = 2
> oninit: Cannot open chunk '/opt/informix/dev/root.1-m'. errno = 2
> oninit: Fatal error in shared memory initialization
>
> The chunks are on a SAN and I use the lvcreate & mknode to create the
> chunks.
> The OS seems to be able to get to the chunk.
>
> # dd if=/opt/informix/dev/root.1 of=/dev/null count=100 bs=1024
> 100+0 records in
> 100+0 records out
>
> Any help would be appreciated.
>
> John
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Art S. Kagel
Oninit (www.oninit.com)
IIUG Board of Directors (art@iiug.org)
Disclaimer: Please keep in mind that my own opinions are my own opinions and
do not reflect on my employer, Oninit, the IIUG, nor any other organization
with which I am associated either explicitly or implicitly. Neither do
those opinions reflect those of other individuals affiliated with any entity
with which I am affiliated nor those of the entities themselves.
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
onint > /var/adm/crash/af.405c0cd > 09:12:30 sane_chopen:pg_addr 0:0 != expected 1:2 > page > page > c00000002a3ef400: 00000000 00000000 00000000 00000000 ........ ........ > .... > 09:12:30 Assert Warning: chunk failed sanity check The above message is saying that the engine is trying to open page 2 (the first checkpoint/Llog reserved page)from chunk #1 and it is getting all 000000 when it is opening it. Something is overwritting your root chunk's reserved pages. You can try this just for grins: dd if=/path/to/rootchunk of=/some/file bs=2048 count=12 (use a 2k block size - same as the page size) And then run od -x on the /some/file You'll likely see a bunch of 000000 on all pages after the onconfig page. That is not an Informix issue... Mike
Also - is ROOTOFFSET set to something other than 0? if it is account for that in your dd cmd. It's either seek or skip - kind of important to figure out which one :-) Mike
Ok, I see a bunch of zeros. Now what? If I understand you email that means an
OS not a Informix problem.
# dd if=/opt/informix/dev/root.1 of=/tmp/jda count=10 bs=2048
10+0 records in
10+0 records out
# od -x /tmp/jda
0000000 0000 0000 0000 0000 0000 0000 0000 0000
*
0050000
Also checked ROOTOFFSET is set to 0
# grep ROOTOFF *onconf.cars:ROOTOFFSET 0 # Offset of root dbspace into device (Kbytes)
>The above message is saying that the engine is trying to open page 2 (the
first checkpoint/Llog reserved >page)from chunk #1 and it is getting all
000000 when it is opening it. Something is overwritting your root >chunk's
reserved pages. You can try this just for grins:
>
>dd if=/path/to/rootchunk of=/some/file bs=2048 count=12 (use a 2k block size
- same as the page size)
>
>And then run od -x on the /some/file
>
>You'll likely see a bunch of 000000 on all pages after the onconfig page.
That is not an Informix issue...
>
>Mike
Well since there is no data there anymore/yet - you can try running oninit -iy
again. Once the engine comes online run your dd command again and see what you
get. Bring the engine down - run dd again, back up, repeat the dd. I dunno
much (anything) about SANS other than the one we use allows other processes
access to many of our disks - meaning some other process could theoretically
overwrite your chunks. Also, back in the day we saw this - when a new raw
device was used you had to provide an offset - small one, like maybe a page or
two - so as to not overwrite the devices VTOC (virtual table of contents).
Like I said this was back in the day so I dunno if that still applies or not.
Mike
Also John - you can, since you are kinda just trying to get the engine up - use a cooked file for your root chunk - just for evidence if you need to get help from Sys Admin re. the SAN issue - or possible SAN issue.
Well I double checked my notes of what I did and it looks like after the
oninit -i I didn't do any dbimports. So I tried one and get this.
dbimport -c -q -i /var/adm/crash -d dbs0 cars
*** create database25588 - The appl process cannot connect to the database server zoom.
2 - No such file or directory
Nothing in the log file, so my guess it's still having problems getting to the
root file. Will see what I can found out on the SAM config.
John
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of MIKE
MAGIE
Sent: Thursday, October 30, 2008 12:44 PM
To: ids@iiug.org
Subject: Re: RE: RE: onint - cannot open chunk error 2 [13854]
Also John - you can, since you are kinda just trying to get the engine up -
use a cooked file for your root chunk - just for evidence if you need to get
help from Sys Admin re. the SAN issue - or possible SAN issue.
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
Try what Mike wrote: make rootdbs cooked file somewhere below INFORMIXDIR.
I do not know about SAN, but you can try to make some offset (ROOTOFFSET).
HTH
Hrvoje
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On
> Behalf Of John Adamski
> Sent: Thursday, October 30, 2008 8:00 PM
> To: ids@iiug.org
> Subject: RE: RE: RE: onint - cannot open chunk error 2 [13856]
>
> Well I double checked my notes of what I did and it looks
> like after the oninit -i I didn't do any dbimports. So I
> tried one and get this.
>
> dbimport -c -q -i /var/adm/crash -d dbs0 cars
> *** create database> 25588 - The appl process cannot connect to the database server zoom.
>
> 2 - No such file or directory
>
> Nothing in the log file, so my guess it's still having
> problems getting to the root file. Will see what I can found
> out on the SAM config.
>
> John
>
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On
> Behalf Of MIKE MAGIE
> Sent: Thursday, October 30, 2008 12:44 PM
> To: ids@iiug.org
> Subject: Re: RE: RE: onint - cannot open chunk error 2 [13854]
>
> Also John - you can, since you are kinda just trying to get
> the engine up - use a cooked file for your root chunk - just
> for evidence if you need to get help from Sys Admin re. the
> SAN issue - or possible SAN issue.
>
>
> **************************************************************
> *****************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
> **************************************************************
> *****************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
Jon - Post or email me a copy of your user informix env., your onconfig file, and your sqlhosts file - are there other informix instances setup on this host? Mike