Can I bypass fast recovery?
Desmodromic had an IDS 9.40.FC8 instance stuck in fast recovery with an assert in rsrecvr.c (Physical log recovery error, Cannot Restore to Checkpoint), no backup, and wanted to bypass fast recovery, avoiding oninit -i. Neil Truby and scottishpoet advised raising an IBM support call since support has tools to bypass it, warning of possible inconsistency requiring integrity checks afterwards. Unresolved.
Auto-generated by Claude from the posts below — may be imperfect; read the full thread.
I have an instance stuck in fast recovery. Here are the last few lines
from the online log:
15:27:27 IBM Informix Dynamic Server Initialized -- Shared MemoryInitialized.
15:27:27 Physical Recovery Started at Page (2:32908).
15:27:27 Assert Failed: Physical log recovery error
15:27:27 IBM Informix Dynamic Server Version 9.40.FC8
15:27:27 Who: Session(12, informix@qroriondb1, 0, 198660378)
Thread(80, fast_rec, 1986362a0, 3)
File: rsrecvr.c Line: 862
15:27:27 Results: Physical Recovery Aborted.
15:27:27 Action: Retry and/or restore
15:27:27 stack trace for pid 3398 written to
/opt/applogs/informix/orion_prod/af.438d0af
15:27:27 See Also: /opt/applogs/informix/orion_prod/af.438d0af,
shmem.438d0af.0
15:29:20 Physical log recovery error
15:29:35 Cannot Restore to Checkpoint.
I am hoping there is some way to bypass fast recovery and bring the
instance online. Does anyone know? No, restore is not an option (I
don't have a backup!). The only other thing I can think of is oninit -i
which I really want to avoid. Any ideas gratefully appreciated.
Thanks.
IBM has some tools reserved for the use of its Tech Support team that might
help you. Raise a support call.
"Desmodromic" <davies_ms@yahoo.com.au> wrote in message
news:1162272061.327795.241990@f16g2000cwb.googlegroups.com...
>I have an instance stuck in fast recovery. Here are the last few lines
> from the online log:
>
> 15:27:27 IBM Informix Dynamic Server Initialized -- Shared Memory> Initialized.
>
> 15:27:27 Physical Recovery Started at Page (2:32908).
> 15:27:27 Assert Failed: Physical log recovery error
> 15:27:27 IBM Informix Dynamic Server Version 9.40.FC8
> 15:27:27 Who: Session(12, informix@qroriondb1, 0, 198660378)
> Thread(80, fast_rec, 1986362a0, 3)
> File: rsrecvr.c Line: 862
> 15:27:27 Results: Physical Recovery Aborted.
> 15:27:27 Action: Retry and/or restore
> 15:27:27 stack trace for pid 3398 written to
> /opt/applogs/informix/orion_prod/af.438d0af
> 15:27:27 See Also: /opt/applogs/informix/orion_prod/af.438d0af,
> shmem.438d0af.0
> 15:29:20 Physical log recovery error
> 15:29:35 Cannot Restore to Checkpoint.>
> I am hoping there is some way to bypass fast recovery and bring the
> instance online. Does anyone know? No, restore is not an option (I
> don't have a backup!). The only other thing I can think of is oninit -i
> which I really want to avoid. Any ideas gratefully appreciated.
>
> Thanks.
>
Desmodromic said: > (I > don't have a backup!). (I don't have any sympathy!) And the period is redundant. -- Bye now, Obnoxio "... no bill is required as no value was provided." -- Christine Normile -- This message has been scanned for viruses and dangerous content by OpenProtect(http://www.openprotect.com), and is believed to be clean.
Call your support provider As the tools support have by pass fast recovery you may end up with corruption and inconsistency in you database as the recovery process will not have completed fully and any uncompleted transactions will neither rollforward nor rollback. So after any activity support has undertaken is completed you will then need to check the integrity of your database. Do you have any transactions that may have been applied to one table and not another ? Then you need to come up with an effective backup and recovery mechanism