Informix will not start
Richard Clark couldn't start IDS 10.00.HC5 on HP-UX: oninit appeared to succeed but no process remained, clients got -25588, and the online log showed every chunk "failed sanity check" and being taken offline, plus assertion failures during fast recovery. Suggestions were to check chunk/symlink ownership and permissions (660, informix-owned, links 777) and the HP-UX volume groups; fixing permissions and recreating links changed nothing. It emerged he was restoring an OS-level make_tape_recovery tape onto a test box, and Everett Mills pointed out such external backups are only valid if the engine was first blocked (onmode -c block). No confirmed fix is recorded in the thread.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
I'm trying to get Informix 10 to start on an HP-UX 11.2 system. The message I
get at boot time is:
Starting Informix OnLine engine(s)
Output from "/sbin/rc3.d/S97informix start":
----------------------------
"/sbin/rc3.d/S97informix start" FAILED
If I run oninit -v everything looks ok (succeeds). But no oninit process shows
up.
If I try to run sql I get the message:
The appl process cannot connect to the database server sabers.
See error number -25588.
Database cars not found or not correct format
onconf:
DBSERVERNAME sabers
sqlhosts:
sabers onipcshm 10.81.101.1 sabers
carsi_tcp onsoctcp 10.81.101.1 istarcarsi
The command dbspace brings up nothing.
Richard Clark
Can you post the last 20 lines or so of the online log - it should have
something in there about the engine starting / failing.
Jarrod Teale
Fonterra NZ
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
RICHARD CLARK
Sent: Friday, 23 January 2009 8:19 a.m.
To: ids@iiug.org
Subject: Informix will not start [14616]
I'm trying to get Informix 10 to start on an HP-UX 11.2 system. The
message I get at boot time is:
Starting Informix OnLine engine(s)
Output from "/sbin/rc3.d/S97informix start":
----------------------------
"/sbin/rc3.d/S97informix start" FAILED
If I run oninit -v everything looks ok (succeeds). But no oninit process
shows up.
If I try to run sql I get the message:
The appl process cannot connect to the database server sabers.
See error number -25588.
Database cars not found or not correct format
onconf:
DBSERVERNAME sabers
sqlhosts:
sabers onipcshm 10.81.101.1 sabers
carsi_tcp onsoctcp 10.81.101.1 istarcarsi
The command dbspace brings up nothing.
Richard Clark
************************************************************************
*******
Forum Note: Use "Reply" to post a response in the discussion forum.
I suspect that they shouldn't have the execute bit set.
Permissions should be 660.
Also the symbolic link isn't owned by informix, so informix can't follow
it to write to the device?
I would:
chmod 777 /opt/informix/dev/dbs0.2
chmod 660 /dev/vg01/rlvol13
Is the volume group okay? I aren't familiar with HPUX volume groups, but
on my systems I would make sure that the volume, logical and physical
volumes are all okay as well.
And try to start again and see if the chuck passes it check.
-----Original Message-----
From: Richard Clark [mailto:clark@salem.edu]
Sent: Friday, 23 January 2009 9:20 a.m.
To: Jarrod Teale
Subject: Re: Informix will not start [14617]
Jarrod,
They're there:
ls -la /opt/informix/dev/dbs0.2
lrwxr-xr-x 1 root common 17 Jan 17 12:24
/opt/informix/dev/dbs0.2@ -> /dev/vg01/rlvol13
ls -la /dev/vg01/rlvol13
crwxrwx--- 1 informix informix 64 0x01000d Jan 17 14:14
/dev/vg01/rlvol13*
No SAN involved.
Richard
Jarrod Teale wrote:
> So are the chunks there? Paths look to be /opt/informix/dev/dbs0.2
> etc...
> Are they permissions 660 and owned by informix.informix?
>
> If you have a SAN behind this box, are the LUN's active?
>
>
>
>
> -----Original Message-----
> From: Richard Clark [mailto:clark@salem.edu]
> Sent: Friday, 23 January 2009 8:47 a.m.
> To: Jarrod Teale
> Subject: Re: Informix will not start [14617]
>
> Jarrod,
>
> It looks like every chunk failed sanity check. Here's the last attempt
> to boot the system:
>
> 13:19:19 IBM Informix Dynamic Server Started.
>
> Thu Jan 22 13:19:20 2009
>
> 13:19:20 Event alarms enabled. ALARMPROG =
> '/opt/informix/etc/no_log.sh'
> 13:19:20 Booting Language <c> from module <> 13:19:20 Loading Module
> <CNULL> 13:19:20 Booting Language <builtin> from module <> 13:19:20
> Loading Module <BUILTINNULL>
> 13:19:26 DR: DRAUTO is 0 (Off)
> 13:19:27 Dynamically allocated new message shared memory segment
> (size
> 2588KB)
> 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5 Software
> Serial Number AAA#B000000
> 13:19:27 Assert Warning: chunk failed sanity check
>
> 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5
> 13:19:27 Who: Session(1, root@sabers, 0, d390c018)
> Thread(17, main_loop(), d38e2018, 1)
> File: rspartn.c Line: 8672
> 13:19:27 Results: Chunk 3 is being taken OFFLINE.
> 13:19:27 Action: Restore chunk from archive.
> 13:19:27 stack trace for pid 9345 written to
> /opt/informix/tmp/af.3f9b8ae
> 13:19:27 See Also: /opt/informix/tmp/af.3f9b8ae
> 13:19:28 chunk failed sanity check
>
> 13:19:28 I/O error, Primary Chunk '/opt/informix/dev/dbs0.2' --
> Offline
> (sanity)
> 13:19:28 Assert Warning: chunk failed sanity check
>
> 13:19:28 IBM Informix Dynamic Server Version 10.00.HC5
> 13:19:28 Who: Session(1, root@sabers, 0, d390c018)
> Thread(17, main_loop(), d38e2018, 1)
> File: rspartn.c Line: 8672
> 13:19:28 Results: Chunk 4 is being taken OFFLINE.
> 13:19:28 Action: Restore chunk from archive.
> 13:19:28 stack trace for pid 9345 written to
> /opt/informix/tmp/af.3f9b8ae
> 13:19:29 See Also: /opt/informix/tmp/af.3f9b8ae
> 13:19:30 chunk failed sanity check
>
> 13:19:30 I/O error, Primary Chunk '/opt/informix/dev/temp0.1' --
> Offline (sanity)
> 13:19:31 Assert Warning: chunk failed sanity check
>
> 13:19:31 IBM Informix Dynamic Server Version 10.00.HC5
> 13:19:31 Who: Session(1, root@sabers, 0, d390c018)
> Thread(17, main_loop(), d38e2018, 1)
> File: rspartn.c Line: 8672
> 13:19:31 Results: Chunk 5 is being taken OFFLINE.
> 13:19:31 Action: Restore chunk from archive.
> 13:19:31 stack trace for pid 9345 written to
> /opt/informix/tmp/af.3f9b8ae
> 13:19:31 See Also: /opt/informix/tmp/af.3f9b8ae
> 13:19:32 chunk failed sanity check
>
> 13:19:33 I/O error, Primary Chunk '/opt/informix/dev/temp1.1' --
> Offline (sanity)
> 13:19:33 Assert Warning: chunk failed sanity check
>
> 13:19:33 IBM Informix Dynamic Server Version 10.00.HC5
> 13:19:33 Who: Session(1, root@sabers, 0, d390c018)
> Thread(17, main_loop(), d38e2018, 1)
> File: rspartn.c Line: 8672
> 13:19:33 Results: Chunk 6 is being taken OFFLINE.
> 13:19:33 Action: Restore chunk from archive.
> 13:19:33 stack trace for pid 9345 written to
> /opt/informix/tmp/af.3f9b8ae
> 13:19:33 See Also: /opt/informix/tmp/af.3f9b8ae
> 13:19:34 chunk failed sanity check
>
> 13:19:35 I/O error, Primary Chunk '/opt/informix/dev/dbs1.1' --
> Offline
> (sanity)
> 13:19:35 Assert Warning: chunk failed sanity check
>
> 13:19:35 IBM Informix Dynamic Server Version 10.00.HC5
> 13:19:35 Who: Session(1, root@sabers, 0, d390c018)
> Thread(17, main_loop(), d38e2018, 1)
> File: rspartn.c Line: 8672
> 13:19:35 Results: Chunk 7 is being taken OFFLINE.
> 13:19:35 Action: Restore chunk from archive.
> 13:19:35 stack trace for pid 9345 written to
> /opt/informix/tmp/af.3f9b8ae
> 13:19:35 See Also: /opt/informix/tmp/af.3f9b8ae
> 13:19:36 chunk failed sanity check
>
> 13:19:37 I/O error, Primary Chunk '/opt/informix/dev/dbs1.2' --
> Offline
> (sanity)
> 13:19:37 IBM Informix Dynamic Server Initialized -- Shared Memory
> Initialized.
>
> 13:19:37 Warning: Invalid (non-existent/blobspace/disabled) dbspace
> listed
> in DBSPACETEMP: 'temp0'
> 13:19:37 Warning: Invalid (non-existent/blobspace/disabled) dbspace
> listed
> in DBSPACETEMP: 'temp1'
> 13:19:37 Physical Recovery Started at Page (1:722).
> 13:19:37 Physical Recovery Complete: 0 Pages Examined, 0 Pages
> Restored.
> 13:19:37 Assert Warning: pthdrpage:ptalloc:bad bfget
> 13:19:37 IBM Informix Dynamic Server Version 10.00.HC5
> 13:19:37 Who: Session(16, root@sabers, 0, d390da80)
> Thread(39, fast_rec, d38e63f0, 1)
> File: rspartn.c Line: 6725
> 13:19:37 Results: Cannot use TBLSpace page for TBLSpace 5242881
> 13:19:37 Action: Run 'oncheck -pt 5242881'
> 13:19:37 stack trace for pid 9345 written to
> /opt/informix/tmp/af.40fb8b9
> 13:19:37 See Also: /opt/informix/tmp/af.40fb8b9
> 13:19:38 pthdrpage:ptalloc:bad bfget
> 13:19:38 Assert Failed: Chunk 6 is being taken OFFLINE.
> 13:19:38 IBM Informix Dynamic Server Version 10.00.HC5
> 13:19:38 Who: Session(16, root@sabers, 0, d390da80)
> Thread(39, fast_rec, d38e63f0, 1)
> File: rspartn.c Line: 6744
> 13:19:38 Results: Cannot Open DBspace 5.
> 13:19:38 Action: Restore chunk from archive.
> in DBSPACETEMP: 'temp1'
> 13:19:37 Physical Recovery Started at Page (1:722).
> 13:19:37 Physical Recovery Complete: 0 Pages Examined, 0 Pages
> Restored.
> 13:19:37 Assert Warning: pthdrpage:ptalloc:bad bfget
> 13:19:37 IBM Informix Dynamic Server Version 10.00.HC5
> 13:19:37 Who: Session(16, root@sabers, 0, d390da80)
> Thread(39, fast_rec, d38e63f0, 1)
> File: rspartn.c Line: 6725
> 13:19:37 Results: Cannot use TBLSpace page for TBLSpace 5242881
> 13:19:37 Action: Run 'oncheck -pt 5242881'
> 13:19:37 stack trace for pid 9345 written to
> /opt/informix/tmp/af.40fb8b9
> 13:19:37 See Also: /opt/informix/tmp/af.40fb8b9
> 13:19:38 pthdrpage:ptalloc:bad bfget
> 13:19:38 Assert Failed: Chunk 6 is being taken OFF
Or at least the link should have 777 permissions set. Root ownership of the
link isn't a problem as IDS does follow the link to the actual chunk
permissions & ownerships if it can, but I suspect that with 755 permissions
on the link and 770 on the underlying chunk IDS will balk. Missed that.
Art
On Thu, Jan 22, 2009 at 3:33 PM, Jarrod Teale <Jarrod.Teale@fonterra.com>wrote:
> I suspect that they shouldn't have the execute bit set.
> Permissions should be 660.
> Also the symbolic link isn't owned by informix, so informix can't follow
> it to write to the device?
>
> I would:
> chmod 777 /opt/informix/dev/dbs0.2
> chmod 660 /dev/vg01/rlvol13
>
> Is the volume group okay? I aren't familiar with HPUX volume groups, but
> on my systems I would make sure that the volume, logical and physical
> volumes are all okay as well.
>
> And try to start again and see if the chuck passes it check.
>
> -----Original Message-----
> From: Richard Clark [mailto:clark@salem.edu]
> Sent: Friday, 23 January 2009 9:20 a.m.
> To: Jarrod Teale
> Subject: Re: Informix will not start [14617]
>
> Jarrod,
>
> They're there:
>
> ls -la /opt/informix/dev/dbs0.2
> lrwxr-xr-x 1 root common 17 Jan 17 12:24
> /opt/informix/dev/dbs0.2@ -> /dev/vg01/rlvol13
>
> ls -la /dev/vg01/rlvol13
> crwxrwx--- 1 informix informix 64 0x01000d Jan 17 14:14
> /dev/vg01/rlvol13*
>
> No SAN involved.
>
> Richard
>
> Jarrod Teale wrote:
> > So are the chunks there? Paths look to be /opt/informix/dev/dbs0.2
> > etc...
> > Are they permissions 660 and owned by informix.informix?
> >
> > If you have a SAN behind this box, are the LUN's active?
> >
> >
> >
> >
> > -----Original Message-----
> > From: Richard Clark [mailto:clark@salem.edu]
> > Sent: Friday, 23 January 2009 8:47 a.m.
> > To: Jarrod Teale
> > Subject: Re: Informix will not start [14617]
> >
> > Jarrod,
> >
> > It looks like every chunk failed sanity check. Here's the last attempt
>
> > to boot the system:
> >
> > 13:19:19 IBM Informix Dynamic Server Started.
> >
> > Thu Jan 22 13:19:20 2009
> >
> > 13:19:20 Event alarms enabled. ALARMPROG =
> > '/opt/informix/etc/no_log.sh'
> > 13:19:20 Booting Language <c> from module <> 13:19:20 Loading Module
>
> > <CNULL> 13:19:20 Booting Language <builtin> from module <> 13:19:20
> > Loading Module <BUILTINNULL>
> > 13:19:26 DR: DRAUTO is 0 (Off)
> > 13:19:27 Dynamically allocated new message shared memory segment
> > (size
> > 2588KB)
> > 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5 Software
> > Serial Number AAA#B000000
> > 13:19:27 Assert Warning: chunk failed sanity check
> >
> > 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5
> > 13:19:27 Who: Session(1, root@sabers, 0, d390c018)
> > Thread(17, main_loop(), d38e2018, 1)
> > File: rspartn.c Line: 8672
> > 13:19:27 Results: Chunk 3 is being taken OFFLINE.
> > 13:19:27 Action: Restore chunk from archive.
> > 13:19:27 stack trace for pid 9345 written to
> > /opt/informix/tmp/af.3f9b8ae
> > 13:19:27 See Also: /opt/informix/tmp/af.3f9b8ae
> > 13:19:28 chunk failed sanity check
> >
> > 13:19:28 I/O error, Primary Chunk '/opt/informix/dev/dbs0.2' --
> > Offline
> > (sanity)
> > 13:19:28 Assert Warning: chunk failed sanity check
> >
> > 13:19:28 IBM Informix Dynamic Server Version 10.00.HC5
> > 13:19:28 Who: Session(1, root@sabers, 0, d390c018)
> > Thread(17, main_loop(), d38e2018, 1)
> > File: rspartn.c Line: 8672
> > 13:19:28 Results: Chunk 4 is being taken OFFLINE.
> > 13:19:28 Action: Restore chunk from archive.
> > 13:19:28 stack trace for pid 9345 written to
> > /opt/informix/tmp/af.3f9b8ae
> > 13:19:29 See Also: /opt/informix/tmp/af.3f9b8ae
> > 13:19:30 chunk failed sanity check
> >
> > 13:19:30 I/O error, Primary Chunk '/opt/informix/dev/temp0.1' --
> > Offline (sanity)
> > 13:19:31 Assert Warning: chunk failed sanity check
> >
> > 13:19:31 IBM Informix Dynamic Server Version 10.00.HC5
> > 13:19:31 Who: Session(1, root@sabers, 0, d390c018)
> > Thread(17, main_loop(), d38e2018, 1)
> > File: rspartn.c Line: 8672
> > 13:19:31 Results: Chunk 5 is being taken OFFLINE.
> > 13:19:31 Action: Restore chunk from archive.
> > 13:19:31 stack trace for pid 9345 written to
> > /opt/informix/tmp/af.3f9b8ae
> > 13:19:31 See Also: /opt/informix/tmp/af.3f9b8ae
> > 13:19:32 chunk failed sanity check
> >
> > 13:19:33 I/O error, Primary Chunk '/opt/informix/dev/temp1.1' --
> > Offline (sanity)
> > 13:19:33 Assert Warning: chunk failed sanity check
> >
> > 13:19:33 IBM Informix Dynamic Server Version 10.00.HC5
> > 13:19:33 Who: Session(1, root@sabers, 0, d390c018)
> > Thread(17, main_loop(), d38e2018, 1)
> > File: rspartn.c Line: 8672
> > 13:19:33 Results: Chunk 6 is being taken OFFLINE.
> > 13:19:33 Action: Restore chunk from archive.
> > 13:19:33 stack trace for pid 9345 written to
> > /opt/informix/tmp/af.3f9b8ae
> > 13:19:33 See Also: /opt/informix/tmp/af.3f9b8ae
> > 13:19:34 chunk failed sanity check
> >
> > 13:19:35 I/O error, Primary Chunk '/opt/informix/dev/dbs1.1' --
> > Offline
> > (sanity)
> > 13:19:35 Assert Warning: chunk failed sanity check
> >
> > 13:19:35 IBM Informix Dynamic Server Version 10.00.HC5
> > 13:19:35 Who: Session(1, root@sabers, 0, d390c018)
> > Thread(17, main_loop(), d38e2018, 1)
> > File: rspartn.c Line: 8672
> > 13:19:35 Results: Chunk 7 is being taken OFFLINE.
> > 13:19:35 Action: Restore chunk from archive.
> > 13:19:35 stack trace for pid 9345 written to
> > /opt/informix/tmp/af.3f9b8ae
> > 13:19:35 See Also: /opt/informix/tmp/af.3f9b8ae
> > 13:19:36 chunk failed sanity check
> >
> > 13:19:37 I/O error, Primary Chunk '/opt/informix/dev/dbs1.2' --
> > Offline
> > (sanity)
> > 13:19:37 IBM Informix Dynamic Server Initialized -- Shared Memory
> > Initialized.
> >
> > 13:19:37 Warning: Invalid (non-existent/blobspace/disabled) dbspace
> > listed
> > in DBSPACETEMP: 'temp0'
> > 13:19:37 Warning: Invalid (non-existent/blobspace/disabled) dbspace
> > listed
> > in DBSPACETEMP: 'temp1'
> > 13:19:37 Physical Recovery Started at Page (1:722).
> > 13:19:37 Physical Recovery Complete: 0 Pages Examined, 0 Pages
> > Restored.
> > 13:19:37 Assert Warning: pthdrpage:ptalloc:bad bfget
> > 13:19:37 IBM Informix Dynamic Server Version 10.00.HC5
> > 13:19:37 Who: Session(16, root@sabers, 0, d390da80)
> > Thread(39, fast_rec, d38e63f0, 1)
> > File: rspartn.c Line: 6725
> > 13:19:37 Results: Cannot use TBLSpace page for TBLSpace 5242881
> > 13:19:37 Action: Run 'oncheck -pt 5242881'
> > 13:19:37 stack trace for pid 9345 written to
> > /opt/informix/tmp/af.40fb8b9
> > 13:19:37 See Also: /opt/informix/tmp/af.40fb8b9
> > 13:19:38 pthdrpage:ptalloc:bad bfget
> > 13:19:38 Assert Failed: Chunk 6 is being taken OFFLINE.
> > 13:19:38 IBM Informix Dynamic Server Version 10.00.HC5
> > 13:19:38 Who: Session(16, root@sabers, 0, d390da80)
> > Thread(39, fast_rec, d38e63f0, 1)
> > File: rspartn.c Line: 6744
> > 13:19:38 Results: Cannot Open DBspace 5.
> > 13:19:38 Action: Restore chunk from archive.
> > in DBSPACET
Do the af files that have been generated show any additional information? Are the errors the same or has it marked the chunks as down and now you need to bring them back online? Also, is this a newly created instance or one that has been running for a while? The next suggestion that I have for you is getting ugly. The assumption is that the chunks still contain Informix data. You could dd out the first couple of pages from the chunks and confirm that they are Informix reserved pages. Art Kagel gave a great presentation about this stuff at the Informix On Demand conference in Las Vegas last year - reading from them is safe - it's when you get to writing them back is when IBM support will tell you to go away -----Original Message----- From: Richard Clark [mailto:clark@salem.edu] Sent: Friday, 23 January 2009 10:34 a.m. To: Jarrod Teale Subject: Re: Informix will not start [14617] I changed the chunks to 660. Logical volumes, volume groups and physical volumes all check out. Informix can follow the link to the device since everyone has an x. Alas, when I rebooted it gives me the same messages. Richard Jarrod Teale wrote: > I suspect that they shouldn't have the execute bit set. > Permissions should be 660. > Also the symbolic link isn't owned by informix, so informix can't > follow it to write to the device? > > I would: > chmod 777 /opt/informix/dev/dbs0.2 > chmod 660 /dev/vg01/rlvol13 > > Is the volume group okay? I aren't familiar with HPUX volume groups, > but on my systems I would make sure that the volume, logical and > physical volumes are all okay as well. > > And try to start again and see if the chuck passes it check. > > -----Original Message----- > From: Richard Clark [mailto:clark@salem.edu] > Sent: Friday, 23 January 2009 9:20 a.m. > To: Jarrod Teale > Subject: Re: Informix will not start [14617] > > Jarrod, > > They're there: > > ls -la /opt/informix/dev/dbs0.2 > lrwxr-xr-x 1 root common 17 Jan 17 12:24 > /opt/informix/dev/dbs0.2@ -> /dev/vg01/rlvol13 > > ls -la /dev/vg01/rlvol13 > crwxrwx--- 1 informix informix 64 0x01000d Jan 17 14:14 > /dev/vg01/rlvol13* > > No SAN involved. > > Richard > > Jarrod Teale wrote: > >> So are the chunks there? Paths look to be /opt/informix/dev/dbs0.2 >> etc... >> Are they permissions 660 and owned by informix.informix? >> >> If you have a SAN behind this box, are the LUN's active? >> >> >> >> >> -----Original Message----- >> From: Richard Clark [mailto:clark@salem.edu] >> Sent: Friday, 23 January 2009 8:47 a.m. >> To: Jarrod Teale >> Subject: Re: Informix will not start [14617] >> >> Jarrod, >> >> It looks like every chunk failed sanity check. Here's the last >> attempt >> > > >> to boot the system: >> >> 13:19:19 IBM Informix Dynamic Server Started. >> >> Thu Jan 22 13:19:20 2009 >> >> 13:19:20 Event alarms enabled. ALARMPROG = >> '/opt/informix/etc/no_log.sh' >> 13:19:20 Booting Language <c> from module <> 13:19:20 Loading >> Module >> > > >> <CNULL> 13:19:20 Booting Language <builtin> from module <> 13:19:20 >> Loading Module <BUILTINNULL> >> 13:19:26 DR: DRAUTO is 0 (Off) >> 13:19:27 Dynamically allocated new message shared memory segment >> (size >> 2588KB) >> 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5 Software >> Serial Number AAA#B000000 >> 13:19:27 Assert Warning: chunk failed sanity check >> >> 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5 >> 13:19:27 Who: Session(1, root@sabers, 0, d390c018) >> Thread(17, main_loop(), d38e2018, 1) >> File: rspartn.c Line: 8672 >> 13:19:27 Results: Chunk 3 is being taken OFFLINE. >> 13:19:27 Action: Restore chunk from archive. >> 13:19:27 stack trace for pid 9345 written to >> /opt/informix/tmp/af.3f9b8ae >> 13:19:27 See Also: /opt/informix/tmp/af.3f9b8ae >> 13:19:28 chunk failed sanity check >> >> 13:19:28 I/O error, Primary Chunk '/opt/informix/dev/dbs0.2' -- >> Offline >> (sanity) >> 13:19:28 Assert Warning: chunk failed sanity check >> >> 13:19:28 IBM Informix Dynamic Server Version 10.00.HC5 >> 13:19:28 Who: Session(1, root@sabers, 0, d390c018) >> Thread(17, main_loop(), d38e2018, 1) >> File: rspartn.c Line: 8672 >> 13:19:28 Results: Chunk 4 is being taken OFFLINE. >> 13:19:28 Action: Restore chunk from archive. >> 13:19:28 stack trace for pid 9345 written to >> /opt/informix/tmp/af.3f9b8ae >> 13:19:29 See Also: /opt/informix/tmp/af.3f9b8ae >> 13:19:30 chunk failed sanity check >> >> 13:19:30 I/O error, Primary Chunk '/opt/informix/dev/temp0.1' -- >> Offline (sanity) >> 13:19:31 Assert Warning: chunk failed sanity check >> >> 13:19:31 IBM Informix Dynamic Server Version 10.00.HC5 >> 13:19:31 Who: Session(1, root@sabers, 0, d390c018) >> Thread(17, main_loop(), d38e2018, 1) >> File: rspartn.c Line: 8672 >> 13:19:31 Results: Chunk 5 is being taken OFFLINE. >> 13:19:31 Action: Restore chunk from archive. >> 13:19:31 stack trace for pid 9345 written to >> /opt/informix/tmp/af.3f9b8ae >> 13:19:31 See Also: /opt/informix/tmp/af.3f9b8ae >> 13:19:32 chunk failed sanity check >> >> 13:19:33 I/O error, Primary Chunk '/opt/informix/dev/temp1.1' -- >> Offline (sanity) >> 13:19:33 Assert Warning: chunk failed sanity check >> >> 13:19:33 IBM Informix Dynamic Server Version 10.00.HC5 >> 13:19:33 Who: Session(1, root@sabers, 0, d390c018) >> Thread(17, main_loop(), d38e2018, 1) >> File: rspartn.c Line: 8672 >> 13:19:33 Results: Chunk 6 is being taken OFFLINE. >> 13:19:33 Action: Restore chunk from archive. >> 13:19:33 stack trace for pid 9345 written to >> /opt/informix/tmp/af.3f9b8ae >> 13:19:33 See Also: /opt/informix/tmp/af.3f9b8ae >> 13:19:34 chunk failed sanity check >> >> 13:19:35 I/O error, Primary Chunk '/opt/informix/dev/dbs1.1' -- >> Offline >> (sanity) >> 13:19:35 Assert Warning: chunk failed sanity check >> >> 13:19:35 IBM Informix Dynamic Server Version 10.00.HC5 >> 13:19:35 Who: Session(1, root@sabers, 0, d390c018) >> Thread(17, main_loop(), d38e2018, 1) >> File: rspartn.c Line: 8672 >> 13:19:35 Results: Chunk 7 is being taken OFFLINE. >> 13:19:35 Action: Restore chunk from archive. >> 13:19:35 stack trace for pid 9345 written to >> /opt/informix/tmp/af.3f9b8ae >> 13:19:35 See Also: /opt/informix/tmp/af.3f9b8ae >> 13:19:36 chunk failed sanity check >> >> 13:19:37 I/O error, Primary Chunk '/opt/informix/dev/dbs1.2' -- >> Offline >> (sanity) >> 13:19:37 IBM Informix Dynamic Server Initialized -- Shared Memory >> Initialized. >> >> 13:19:37 Warning: Invalid (non-existent/blobspace/disabled) dbspace >> listed >> in DBSPACETEMP: 'temp0' >> 13:19:37 Warning: Invalid (non-existent/blobspace/disabled) dbspace >> listed >> in DBSPACETEMP: 'temp1' >> 13:19:37 Physical Recovery Started at Page (1:722). >> 13:19:37 Physical Recovery Complete: 0 Pages Examined, 0 Pages >> Restored. >> 13:19:37 Assert Warning: pthdrpage:ptalloc:bad bfget >> 13:19:37 IBM Informix Dynamic Server Version 10.00.HC5 >> 13:19:37 Who: Session(16, root@sabers, 0, d390da80)@
I recreated the links as user informix and rebooted. No help. I'm not familiar with the dd process but I used it to create a file that obviously had the data from my database in it. Not sure what that means in relation to what you're saying about reserved pages. What I'm doing is recovering with a tape (make_tape_recovery -I -v -x inc_entire=vg00 -x inc_entire=vg01 -x inc_entire=vg02) to an identical test box. This has worked several times in the past year but I've missed something this time. The af files look like they are taking the chunks down and bringing them back: 17:06:29 Assert Warning: chunk failed sanity check 17:06:29 Who: Session(1, root@sabers, 0, c80e5018) Thread(17, main_loop(), c80bb018, 3) File: rspartn.c Line: 8672 17:06:29 Results: Chunk 3 is being taken OFFLINE. 17:06:29 Action: Restore chunk from archive. 17:06:29 Stack for thread: 17 main_loop() base: 0xc8790000 len: 69632 pc: 0x00000000 tos: 0xc8791e90 state: running vp: 3 ( 0) 0x008a7a1c legacy_hp_afstack + 0x198 [/opt/informix/bin/oninit] ( 1) 0x008a72c0 afstack + 0x48 [/opt/informix/bin/oninit] ( 2) 0x008a668c afhandler + 0x8a0 [/opt/informix/bin/oninit] ( 3) 0x008a5d4c afwarn_interface + 0x48 [/opt/informix/bin/oninit] ( 4) 0x005b6f88 sane_chopen + 0x240 [/opt/informix/bin/oninit] ( 5) 0x005b6950 chopen_util + 0x1cc [/opt/informix/bin/oninit] ( 6) 0x005b66e0 chopen + 0x10 [/opt/informix/bin/oninit] ( 7) 0x0012244c initchunk + 0x1d8 [/opt/informix/bin/oninit] ( 8) 0x0011eaa8 initseg + 0xbf4 [/opt/informix/bin/oninit] ( 9) 0x00113ed4 main_loop + 0x534 [/opt/informix/bin/oninit] (10) 0x008af9e4 startup + 0x94 [/opt/informix/bin/oninit] (11) 0x008847e0 mt_swap_threads + 0x10c [/opt/informix/bin/oninit] base: 0xc8790000 len: 69632 pc: 0x00000000 tos: 0xc8791e90 state: running vp: 3 17:06:29 See Also: /opt/informix/tmp/af.3f9ede4 --------------------------------- Begin System Alarm Program Output --------------------------------- Assertion Failure Type: Warning Host Name: sabers Database Server Name: sabers Time of failure: Thu Jan 22 17:06:30 EST 2009 AF file: /opt/informix/tmp/af.3f9ede4 Shared memory file: None System Blocking: OFF ------------------------------- End System Alarm Program Output ------------------------------- 17:06:30 sh /opt/informix/etc/evidence.sh 1 0 /opt/informix/tmp/af.3f9ede4 1 0xc80bb018 17 0xc8786e00 1025 0 0 0 0 17:06:30 ------------------ End of assertion failure 0 ----------------- Jarrod Teale wrote: > Do the af files that have been generated show any additional > information? > Are the errors the same or has it marked the chunks as down and now you > need to bring them back online? > Also, is this a newly created instance or one that has been running for > a while? > > The next suggestion that I have for you is getting ugly. The assumption > is that the chunks still contain Informix data. You could dd out the > first couple of pages from the chunks and confirm that they are Informix > reserved pages. Art Kagel gave a great presentation about this stuff at > the Informix On Demand conference in Las Vegas last year - reading from > them is safe - it's when you get to writing them back is when IBM > support will tell you to go away > > -----Original Message----- > From: Richard Clark [mailto:clark@salem.edu] > Sent: Friday, 23 January 2009 10:34 a.m. > To: Jarrod Teale > Subject: Re: Informix will not start [14617] > > I changed the chunks to 660. Logical volumes, volume groups and physical > volumes all check out. Informix can follow the link to the device since > everyone has an x. > > Alas, when I rebooted it gives me the same messages. > > Richard > > Jarrod Teale wrote: > >> I suspect that they shouldn't have the execute bit set. >> Permissions should be 660. >> Also the symbolic link isn't owned by informix, so informix can't >> follow it to write to the device? >> >> I would: >> chmod 777 /opt/informix/dev/dbs0.2 >> chmod 660 /dev/vg01/rlvol13 >> >> Is the volume group okay? I aren't familiar with HPUX volume groups, >> but on my systems I would make sure that the volume, logical and >> physical volumes are all okay as well. >> >> And try to start again and see if the chuck passes it check. >> >> -----Original Message----- >> From: Richard Clark [mailto:clark@salem.edu] >> Sent: Friday, 23 January 2009 9:20 a.m. >> To: Jarrod Teale >> Subject: Re: Informix will not start [14617] >> >> Jarrod, >> >> They're there: >> >> ls -la /opt/informix/dev/dbs0.2 >> lrwxr-xr-x 1 root common 17 Jan 17 12:24 >> /opt/informix/dev/dbs0.2@ -> /dev/vg01/rlvol13 >> >> ls -la /dev/vg01/rlvol13 >> crwxrwx--- 1 informix informix 64 0x01000d Jan 17 14:14 >> /dev/vg01/rlvol13* >> >> No SAN involved. >> >> Richard >> >> Jarrod Teale wrote: >> >> >>> So are the chunks there? Paths look to be /opt/informix/dev/dbs0.2 >>> etc... >>> Are they permissions 660 and owned by informix.informix? >>> >>> If you have a SAN behind this box, are the LUN's active? >>> >>> >>> >>> >>> -----Original Message----- >>> From: Richard Clark [mailto:clark@salem.edu] >>> Sent: Friday, 23 January 2009 8:47 a.m. >>> To: Jarrod Teale >>> Subject: Re: Informix will not start [14617] >>> >>> Jarrod, >>> >>> It looks like every chunk failed sanity check. Here's the last >>> attempt >>> >>> >> >>> to boot the system: >>> >>> 13:19:19 IBM Informix Dynamic Server Started. >>> >>> Thu Jan 22 13:19:20 2009 >>> >>> 13:19:20 Event alarms enabled. ALARMPROG = >>> '/opt/informix/etc/no_log.sh' >>> 13:19:20 Booting Language <c> from module <> 13:19:20 Loading >>> Module >>> >>> >> >>> <CNULL> 13:19:20 Booting Language <builtin> from module <> 13:19:20 >>> Loading Module <BUILTINNULL> >>> 13:19:26 DR: DRAUTO is 0 (Off) >>> 13:19:27 Dynamically allocated new message shared memory segment >>> (size >>> 2588KB) >>> 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5 Software >>> Serial Number AAA#B000000 >>> 13:19:27 Assert Warning: chunk failed sanity check >>> >>> 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5 >>> 13:19:27 Who: Session(1, root@sabers, 0, d390c018) >>> Thread(17, main_loop(), d38e2018, 1) >>> File: rspartn.c Line: 8672 >>> 13:19:27 Results: Chunk 3 is being taken OFFLINE. >>> 13:19:27 Action: Restore chunk from archive. >>> 13:19:27 stack trace for pid 9345 written to >>> /opt/informix/tmp/af.3f9b8ae >>> 13:19:27 See Also: /opt/informix/tmp/af.3f9b8ae >>> 13:19:28 chunk failed sanity check >>> >>> 13:19:28 I/O error, Primary Chunk '/opt/informix/dev/dbs0.2' -- >>> Offline >>> (sanity) >>> 13:19:28 Assert Warning: chunk failed sanity check >>> >>> 13:19:28 IBM Informix Dynamic Server Version 10.00.HC5 >>> 13:19:28 Who: Session(1, root@sabers, 0, d390c018) >>> Thread(17, main_loop(), d38e2018, 1) >>> File: rspartn.c Line: 8672 >>> 13:19:28 Results: Chunk 4 is being taken OFFLINE. >>> 13:19:28 Action:
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> Richard Clark
> Sent: Thursday, January 22, 2009 4:46 PM
> To: ids@iiug.org
> Subject: Re: Informix will not start [14625]
>
> I recreated the links as user informix and rebooted. No help.
>
> I'm not familiar with the dd process but I used it to create a file
that
> obviously had the data from my database in it. Not sure what that
means
> in relation to what you're saying about reserved pages.
>
> What I'm doing is recovering with a tape (make_tape_recovery -I -v -x
> inc_entire=vg00 -x inc_entire=vg01 -x inc_entire=vg02) to an identical
> test box. This has worked several times in the past year but I've
missed
> something this time.
>
> The af files look like they are taking the chunks down and bringing
them
> back:
>
> 17:06:29 Assert Warning: chunk failed sanity check
>
> 17:06:29 Who: Session(1, root@sabers, 0, c80e5018)
>
> Thread(17, main_loop(), c80bb018, 3)
>
> File: rspartn.c Line: 8672
> 17:06:29 Results: Chunk 3 is being taken OFFLINE.
> 17:06:29 Action: Restore chunk from archive.
> 17:06:29 Stack for thread: 17 main_loop()
You are obviously restoring a tape made outside of the database
engine. This is only supported if the engine is put into outside backup
mode (onmode -c block) to stop transactions from being written to the
disk during your backup.
Did you block the server before running the archive?
--EEM
>
> base: 0xc8790000
> len: 69632
>
> pc: 0x00000000
> tos: 0xc8791e90
> state: running
>
> vp: 3
>
> ( 0) 0x008a7a1c legacy_hp_afstack + 0x198 [/opt/informix/bin/oninit]
> ( 1) 0x008a72c0 afstack + 0x48 [/opt/informix/bin/oninit]
> ( 2) 0x008a668c afhandler + 0x8a0 [/opt/informix/bin/oninit]
> ( 3) 0x008a5d4c afwarn_interface + 0x48 [/opt/informix/bin/oninit]
> ( 4) 0x005b6f88 sane_chopen + 0x240 [/opt/informix/bin/oninit]
> ( 5) 0x005b6950 chopen_util + 0x1cc [/opt/informix/bin/oninit]
> ( 6) 0x005b66e0 chopen + 0x10 [/opt/informix/bin/oninit]
> ( 7) 0x0012244c initchunk + 0x1d8 [/opt/informix/bin/oninit]
> ( 8) 0x0011eaa8 initseg + 0xbf4 [/opt/informix/bin/oninit]
> ( 9) 0x00113ed4 main_loop + 0x534 [/opt/informix/bin/oninit]
> (10) 0x008af9e4 startup + 0x94 [/opt/informix/bin/oninit]
> (11) 0x008847e0 mt_swap_threads + 0x10c [/opt/informix/bin/oninit]
>
> base: 0xc8790000
> len: 69632
>
> pc: 0x00000000
> tos: 0xc8791e90
> state: running
>
> vp: 3
>
> 17:06:29 See Also: /opt/informix/tmp/af.3f9ede4
>
> ---------------------------------
> Begin System Alarm Program Output
> ---------------------------------
>
> Assertion Failure Type: Warning
> Host Name: sabers
> Database Server Name: sabers
> Time of failure: Thu Jan 22 17:06:30 EST 2009
> AF file: /opt/informix/tmp/af.3f9ede4
> Shared memory file: None
> System Blocking: OFF
>
> -------------------------------
> End System Alarm Program Output
> -------------------------------
>
> 17:06:30 sh /opt/informix/etc/evidence.sh 1 0
> /opt/informix/tmp/af.3f9ede4 1 0xc80bb018 17 0xc8786e00 1025 0 0 0 0
> 17:06:30
> ------------------ End of assertion failure 0 -----------------
>
> Jarrod Teale wrote:
> > Do the af files that have been generated show any additional
> > information?
> > Are the errors the same or has it marked the chunks as down and now
you
> > need to bring them back online?
> > Also, is this a newly created instance or one that has been running
for
> > a while?
> >
> > The next suggestion that I have for you is getting ugly. The
assumption
> > is that the chunks still contain Informix data. You could dd out the
> > first couple of pages from the chunks and confirm that they are
Informix
> > reserved pages. Art Kagel gave a great presentation about this stuff
at
> > the Informix On Demand conference in Las Vegas last year - reading
from
> > them is safe - it's when you get to writing them back is when IBM
> > support will tell you to go away
> >
> > -----Original Message-----
> > From: Richard Clark [mailto:clark@salem.edu]
> > Sent: Friday, 23 January 2009 10:34 a.m.
> > To: Jarrod Teale
> > Subject: Re: Informix will not start [14617]
> >
> > I changed the chunks to 660. Logical volumes, volume groups and
physical
> > volumes all check out. Informix can follow the link to the device
since
> > everyone has an x.
> >
> > Alas, when I rebooted it gives me the same messages.
> >
> > Richard
> >
> > Jarrod Teale wrote:
> >
> >> I suspect that they shouldn't have the execute bit set.
> >> Permissions should be 660.
> >> Also the symbolic link isn't owned by informix, so informix can't
> >> follow it to write to the device?
> >>
> >> I would:
> >> chmod 777 /opt/informix/dev/dbs0.2
> >> chmod 660 /dev/vg01/rlvol13
> >>
> >> Is the volume group okay? I aren't familiar with HPUX volume
groups,
> >> but on my systems I would make sure that the volume, logical and
> >> physical volumes are all okay as well.
> >>
> >> And try to start again and see if the chuck passes it check.
> >>
> >> -----Original Message-----
> >> From: Richard Clark [mailto:clark@salem.edu]
> >> Sent: Friday, 23 January 2009 9:20 a.m.
> >> To: Jarrod Teale
> >> Subject: Re: Informix will not start [14617]
> >>
> >> Jarrod,
> >>
> >> They're there:
> >>
> >> ls -la /opt/informix/dev/dbs0.2
> >> lrwxr-xr-x 1 root common 17 Jan 17 12:24
> >> /opt/informix/dev/dbs0.2@ -> /dev/vg01/rlvol13
> >>
> >> ls -la /dev/vg01/rlvol13
> >> crwxrwx--- 1 informix informix 64 0x01000d Jan 17 14:14
> >> /dev/vg01/rlvol13*
> >>
> >> No SAN involved.
> >>
> >> Richard
> >>
> >> Jarrod Teale wrote:
> >>
> >>
> >>> So are the chunks there? Paths look to be /opt/informix/dev/dbs0.2
> >>> etc...
> >>> Are they permissions 660 and owned by informix.informix?
> >>>
> >>> If you have a SAN behind this box, are the LUN's active?
> >>>
> >>>
> >>>
> >>>
> >>> -----Original Message-----
> >>> From: Richard Clark [mailto:clark@salem.edu]
> >>> Sent: Friday, 23 January 2009 8:47 a.m.
> >>> To: Jarrod Teale
> >>> Subject: Re: Informix will not start [14617]
> >>>
> >>> Jarrod,
> >>>
> >>> It looks like every chunk failed sanity check. Here's the last
> >>> attempt
> >>>
> >>>
> >>
> >>> to boot the system:
> >>>
> >>> 13:19:19 IBM Informix Dynamic Server Started.
> >>>
> >>> Thu Jan 22 13:19:20 2009
> >>>
> >>> 13:19:20 Event alarms enabled. ALARMPROG =
> >>> '/opt/informix/etc/no_log.sh'
> >>> 13:19:20 Booting Language <c> from module <> 13:19:20 Loading
> >>> Module
> >>>
> >>>
> >>
> >>> <CNULL> 13:19:20 Booting Language <builtin> from module <>
13:19:20
> >>> Loading Module <BUILTINNULL>
> >>> 13:19:26 DR: DRAUTO is 0 (Off)
> >>> 13:19:27 Dynamically allocated new message shared memory segment
> >>> (size
> >>> 2588KB)
> >>> 13:19:27 IBM Informix Dynamic Server Version 10.00.HC5 Software
> >>> Serial Number AAA#B000000
> >>> 13:19:27
> No I didn't. I've never heard that before. Are you saying that if I
> should block the server now and proceed with restoring the database
> from daily backup tapes?
>
> Richard
The process is:
On the live server: onmode -c block
Run your backup
On the live server: onmode -c unblock
Then you can restore the archive you just made to the target computer.
THIS WILL PREVENT YOUR SERVER FROM ISSUEING A CHECKPOINT UNTIL THE
BACKUP COMPLETES. ACCORDING TO THE MANUAL, USERS WILL BE IN READ ONLY
MODE (I've never used it, myself).
It is much better (and easier) to use the online backup utilities
provided with the database engine, ontape and onbar. If you need to
have your archives all on a single tape, you can make your backup go to
a disk file which your other backup process can pick up. Ontape can be
configured with just a few onconfig params and the above process is
reduced to this:
On the live server: ontape -s -L 0
On the target server: ontape -r
--EEM
>
> >> -----Original Message-----
> >> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf
Of
> >> Richard Clark
> >> Sent: Thursday, January 22, 2009 4:46 PM
> >> To: ids@iiug.org
> >> Subject: Re: Informix will not start [14625]
> >>
> >> I recreated the links as user informix and rebooted. No help.
> >>
> >> I'm not familiar with the dd process but I used it to create a file
> > that
> >> obviously had the data from my database in it. Not sure what that
> > means
> >> in relation to what you're saying about reserved pages.
> >>
> >> What I'm doing is recovering with a tape (make_tape_recovery -I -v
-x
> >> inc_entire=vg00 -x inc_entire=vg01 -x inc_entire=vg02) to an
identical
> >> test box. This has worked several times in the past year but I've
> > missed
> >> something this time.
> >>
> >> The af files look like they are taking the chunks down and bringing
> > them
> >> back:
> >>
> >> 17:06:29 Assert Warning: chunk failed sanity check
> >>
> >> 17:06:29 Who: Session(1, root@sabers, 0, c80e5018)
> >>
> >> Thread(17, main_loop(), c80bb018, 3)
> >>
> >> File: rspartn.c Line: 8672
> >> 17:06:29 Results: Chunk 3 is being taken OFFLINE.
> >> 17:06:29 Action: Restore chunk from archive.
> >> 17:06:29 Stack for thread: 17 main_loop()
> >
> > You are obviously restoring a tape made outside of the database
> > engine. This is only supported if the engine is put into outside
backup
> > mode (onmode -c block) to stop transactions from being written to
the
> > disk during your backup.
> >
> > Did you block the server before running the archive?
> >
> > --EEM
> >>
> >> base: 0xc8790000
> >> len: 69632
> >>
> >> pc: 0x00000000
> >> tos: 0xc8791e90
> >> state: running
> >>
> >> vp: 3
> >>
> >> ( 0) 0x008a7a1c legacy_hp_afstack + 0x198
[/opt/informix/bin/oninit]
> >> ( 1) 0x008a72c0 afstack + 0x48 [/opt/informix/bin/oninit]
> >> ( 2) 0x008a668c afhandler + 0x8a0 [/opt/informix/bin/oninit]
> >> ( 3) 0x008a5d4c afwarn_interface + 0x48 [/opt/informix/bin/oninit]
> >> ( 4) 0x005b6f88 sane_chopen + 0x240 [/opt/informix/bin/oninit]
> >> ( 5) 0x005b6950 chopen_util + 0x1cc [/opt/informix/bin/oninit]
> >> ( 6) 0x005b66e0 chopen + 0x10 [/opt/informix/bin/oninit]
> >> ( 7) 0x0012244c initchunk + 0x1d8 [/opt/informix/bin/oninit]
> >> ( 8) 0x0011eaa8 initseg + 0xbf4 [/opt/informix/bin/oninit]
> >> ( 9) 0x00113ed4 main_loop + 0x534 [/opt/informix/bin/oninit]
> >> (10) 0x008af9e4 startup + 0x94 [/opt/informix/bin/oninit]
> >> (11) 0x008847e0 mt_swap_threads + 0x10c [/opt/informix/bin/oninit]
> >>
> >> base: 0xc8790000
> >> len: 69632
> >>
> >> pc: 0x00000000
> >> tos: 0xc8791e90
> >> state: running
> >>
> >> vp: 3
> >>
> >> 17:06:29 See Also: /opt/informix/tmp/af.3f9ede4
> >>
> >> ---------------------------------
> >> Begin System Alarm Program Output
> >> ---------------------------------
> >>
> >> Assertion Failure Type: Warning
> >> Host Name: sabers
> >> Database Server Name: sabers
> >> Time of failure: Thu Jan 22 17:06:30 EST 2009
> >> AF file: /opt/informix/tmp/af.3f9ede4
> >> Shared memory file: None
> >> System Blocking: OFF
> >>
> >> -------------------------------
> >> End System Alarm Program Output
> >> -------------------------------
> >>
> >> 17:06:30 sh /opt/informix/etc/evidence.sh 1 0
> >> /opt/informix/tmp/af.3f9ede4 1 0xc80bb018 17 0xc8786e00 1025 0 0 0
0
> >> 17:06:30
> >> ------------------ End of assertion failure 0 -----------------
> >>
> >> Jarrod Teale wrote:
> >> > Do the af files that have been generated show any additional
> >> > information?
> >> > Are the errors the same or has it marked the chunks as down and
now
> > you
> >> > need to bring them back online?
> >> > Also, is this a newly created instance or one that has been
running
> > for
> >> > a while?
> >> >
> >> > The next suggestion that I have for you is getting ugly. The
> > assumption
> >> > is that the chunks still contain Informix data. You could dd out
the
> >> > first couple of pages from the chunks and confirm that they are
> > Informix
> >> > reserved pages. Art Kagel gave a great presentation about this
stuff
> > at
> >> > the Informix On Demand conference in Las Vegas last year -
reading
> > from
> >> > them is safe - it's when you get to writing them back is when IBM
> >> > support will tell you to go away
> >> >
> >> > -----Original Message-----
> >> > From: Richard Clark [mailto:clark@salem.edu]
> >> > Sent: Friday, 23 January 2009 10:34 a.m.
> >> > To: Jarrod Teale
> >> > Subject: Re: Informix will not start [14617]
> >> >
> >> > I changed the chunks to 660. Logical volumes, volume groups and
> > physical
> >> > volumes all check out. Informix can follow the link to the device
> > since
> >> > everyone has an x.
> >> >
> >> > Alas, when I rebooted it gives me the same messages.
> >> >
> >> > Richard
> >> >
> >> > Jarrod Teale wrote:
> >> >
> >> >> I suspect that they shouldn't have the execute bit set.
> >> >> Permissions should be 660.
> >> >> Also the symbolic link isn't owned by informix, so informix
can't
> >> >> follow it to write to the device?
> >> >>
> >> >> I would:
> >> >> chmod 777 /opt/informix/dev/dbs0.2
> >> >> chmod 660 /dev/vg01/rlvol13
> >> >>
> >> >> Is the volume group okay? I aren't familiar with HPUX volume
> > groups,
> >> >> but on my systems I would make sure that the volume, logical and
> >> >> physical volumes are all okay as well.
> >> >>
> >> >> And try to start again and see if the chuck passes it check.
> >> >>
> >> >> -----Original Message-----
> >> >> From: Richard Clark [mailto:clark@salem.edu]
> >> >> Sent: Friday, 23 January 2009 9:20 a.m.
> >> >> To: Jarrod Teale
> >> >> Subject: Re: Informix will not start [14617]
> >> >>
> >> >> Jarrod,
> >> >>
> >> >> They're there:
> >> >>
> >> >> ls -la /opt/informix/dev/dbs0.2
> >> >> lrwxr-xr-x 1 root common 17 Jan 17 12:24
> >> >> /opt/informix/dev/dbs0.2@ -> /dev/vg01/rlvol
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