Onbar PIT restore problem/question

Any ideas would be appreciated! I perform the following everyday at 22:00. onbar -b -L 0 -w It takes approx. 40 minutes to run. I got a request to perform a restore. I was to restore the instance to 22:38:22 the night before. I then issued the following command. onbar -r -t 1999-08-31 22:38:22 The log number which contained this time was 5711. While monitoring the restore I noticed the following message in the online log. PIT reached - logid: 5711, logpos: 0x2e1158 I thought the restore would be completed soon. However, the restore continued for many hours. The restore was going through all the logs since 5711. I finally killed the restore. I then performed another restore. However, this time I chose to restore to the log number 5710. This worked perfectly. My environment is as follows: Informix 7.24.UCX1, HP Omniback II Version A.03.00, HP-UX 10.20 Thanks for any ideas! G Sent via Deja.com http://www.deja.com/ Share what you know. Learn what you don't.

Hmmm .... i don't have any experiences with whole system backups (AFAIK they are only useful, if you do not back up your logical logs), but i think you have to use onbar -r -w -t 1999-08-31 22:38:22 because u use -w for backup also (try "onbar -?"). Actually i experienced many issues with onbar restores, especially with PIT ... after some successful tests i thought it will work in the future, but some months later it did not work anymore (neither -t nor -n), so we finally edited the ixbar file (we deleted all lines which contained logfiles we did not want to be restored) and performed an onbar -r ... this worked. BTW: Maybe your 7.24 has a bug also, the Readme for Networker on Reliant Unix (Siemens RM machines) says: D) NSR-DBM-INF with Informix version 7.24 An Informix error causes the database to be made inconsistent if 'onbar -r' is used. Hope i could give you some hints. Wolf PS.: Tell me the solution, if there is one. fitr0912@my-deja.com wrote: > Any ideas would be appreciated! > > I perform the following everyday at 22:00. > > onbar -b -L 0 -w> > It takes approx. 40 minutes to run. I got a request to perform a > restore. I was to restore the instance to 22:38:22 the night before. I > then issued the following command. > > onbar -r -t 1999-08-31 22:38:22> > The log number which contained this time was 5711. While monitoring the > restore I noticed the following message in the online log. > > PIT reached - logid: 5711, logpos: 0x2e1158 > > I thought the restore would be completed soon. However, the restore > continued for many hours. The restore was going through all the logs > since 5711. I finally killed the restore. I then performed another > restore. However, this time I chose to restore to the log number 5710. > This worked perfectly. > > My environment is as follows: Informix 7.24.UCX1, HP Omniback II Version > A.03.00, HP-UX 10.20 > > Thanks for any ideas! > > G > > Sent via Deja.com http://www.deja.com/ > Share what you know. Learn what you don't. -- Remove the uppercase words from my e-mail address to reply.

You probably ran into this bug# Bug: 74538 ONBAR POINT IN TIME RECOVERY DOES NOT STOP DATA RECOVERY AT TIME SPECIFIED Description: OnBar Point-In-Time Recovery restores data beyond the time specified. Similiar to Bug #74370 for XPS and the same test can be used without the xps commands xcntl for verification. Bug Fixed: 7.30.UC1. --- Bug: 79270 GLS/9.12/9.2 PIT RESTORE RECEIVES SIGNAL 10 OR SIGNAL 11 Description: 1997-07-31 11:04:26 11587 11580 /vobs/tristarp/sqldist/bin/onbar -b 1997-07-31 11:04:32 11587 11580 Begin level 0 backup rootdbs. 1997-07-31 11:04:33 11587 11580 Successfully connected to Storage Manager. 1997-07-31 11:04:34 11587 11580 Completed level 0 backup rootdbs. 1997-07-31 11:04:35 11587 11580 Begin level 0 backup blobspace1. 1997-07-31 11:04:35 11587 11580 Completed level 0 backup blobspace1. 1997-07-31 11:04:35 11587 11580 Begin level 0 backup blobspace2. 1997-07-31 11:04:35 11587 11580 Completed level 0 backup blobspace2. 1997-07-31 11:04:36 11587 11580 Begin level 0 backup blobspace3. 1997-07-31 11:04:36 11587 11580 Completed level 0 backup blobspace3. 1997-07-31 11:04:36 11587 11580 Begin level 0 backup dbspace1. 1997-07-31 11:04:37 11587 11580 Completed level 0 backup dbspace1. 1997-07-31 11:04:37 11587 11580 Begin level 0 backup dbspace2. 1997-07-31 11:04:37 11587 11580 Completed level 0 backup dbspace2. 1997-07-31 11:04:37 11587 11580 Begin level 0 backup dbspace3. 1997-07-31 11:04:38 11587 11580 Completed level 0 backup dbspace3. 1997-07-31 11:04:40 11587 11580 Begin backup logical log 1. 1997-07-31 11:04:40 11587 11580 Successfully connected to Storage Manager. 1997-07-31 11:04:41 11587 11580 Completed backup logical log 1. 1997-07-31 11:04:41 11587 11580 Begin backup logical log 2. 1997-07-31 11:04:41 11587 11580 Completed backup logical log 2. 1997-07-31 11:04:41 11587 11580 Begin backup logical log 3. 1997-07-31 11:04:43 11587 11580 Completed backup logical log 3. 1997-07-31 11:05:13 11670 11661 /vobs/tristarp/sqldist/bin/onbar -l -c 1997-07-31 11:05:14 11670 11661 Begin backup logical log 4. 1997-07-31 11:05:14 11670 11661 Successfully connected to Storage Manager. 1997-07-31 11:05:14 11670 11661 Completed backup logical log 4. 1997-07-31 11:07:03 12094 12079 /vobs/tristarp/sqldist/bin/onbar -r -t 1997\\170#^H-07\\170#^H-31\\170#^H 11\\170# ^H:04\\170#^H:51 --Anil fitr0912@my-deja.com wrote: > Any ideas would be appreciated! > > I perform the following everyday at 22:00. > > onbar -b -L 0 -w> > It takes approx. 40 minutes to run. I got a request to perform a > restore. I was to restore the instance to 22:38:22 the night before. I > then issued the following command. > > onbar -r -t 1999-08-31 22:38:22> > The log number which contained this time was 5711. While monitoring the > restore I noticed the following message in the online log. > > PIT reached - logid: 5711, logpos: 0x2e1158 > > I thought the restore would be completed soon. However, the restore > continued for many hours. The restore was going through all the logs > since 5711. I finally killed the restore. I then performed another > restore. However, this time I chose to restore to the log number 5710. > This worked perfectly. > > My environment is as follows: Informix 7.24.UCX1, HP Omniback II Version > A.03.00, HP-UX 10.20 > > Thanks for any ideas! > > G > > Sent via Deja.com http://www.deja.com/ > Share what you know. Learn what you don't.

Anil, Thanks for the response. However, my best guess is that Informix was not actually restoring the data beyond the PIT. I think it was trying to resolve any open transactions at the PIT. Another person has informed me of bug #107710. What do you think. Thanks, G In article <37D5A60D.CA44FEC@informix.com>, Anil Karkera <anilk@informix.com> wrote: > You probably ran into this bug# > > Bug: 74538 ONBAR POINT IN TIME RECOVERY DOES NOT > STOP DATA RECOVERY AT TIME SPECIFIED > > Description: > OnBar Point-In-Time Recovery restores data beyond the time specified. > Similiar to Bug #74370 for XPS and the same test can be used without > the xps commands xcntl for verification. > > Bug Fixed: 7.30.UC1. > > --- > > Bug: 79270 GLS/9.12/9.2 PIT RESTORE RECEIVES SIGNAL 10 OR SIGNAL 11 > > Description: > 1997-07-31 11:04:26 11587 11580 /vobs/tristarp/sqldist/bin/onbar -b > 1997-07-31 11:04:32 11587 11580 Begin level 0 backup rootdbs. > 1997-07-31 11:04:33 11587 11580 Successfully connected to Storage Manager. > > 1997-07-31 11:04:34 11587 11580 Completed level 0 backup rootdbs. > 1997-07-31 11:04:35 11587 11580 Begin level 0 backup blobspace1. > 1997-07-31 11:04:35 11587 11580 Completed level 0 backup blobspace1. > 1997-07-31 11:04:35 11587 11580 Begin level 0 backup blobspace2. > 1997-07-31 11:04:35 11587 11580 Completed level 0 backup blobspace2. > 1997-07-31 11:04:36 11587 11580 Begin level 0 backup blobspace3. > 1997-07-31 11:04:36 11587 11580 Completed level 0 backup blobspace3. > 1997-07-31 11:04:36 11587 11580 Begin level 0 backup dbspace1. > 1997-07-31 11:04:37 11587 11580 Completed level 0 backup dbspace1. > 1997-07-31 11:04:37 11587 11580 Begin level 0 backup dbspace2. > 1997-07-31 11:04:37 11587 11580 Completed level 0 backup dbspace2. > 1997-07-31 11:04:37 11587 11580 Begin level 0 backup dbspace3. > 1997-07-31 11:04:38 11587 11580 Completed level 0 backup dbspace3. > 1997-07-31 11:04:40 11587 11580 Begin backup logical log 1. > 1997-07-31 11:04:40 11587 11580 Successfully connected to Storage Manager. > > 1997-07-31 11:04:41 11587 11580 Completed backup logical log 1. > 1997-07-31 11:04:41 11587 11580 Begin backup logical log 2. > 1997-07-31 11:04:41 11587 11580 Completed backup logical log 2. > 1997-07-31 11:04:41 11587 11580 Begin backup logical log 3. > 1997-07-31 11:04:43 11587 11580 Completed backup logical log 3. > 1997-07-31 11:05:13 11670 11661 /vobs/tristarp/sqldist/bin/onbar -l -c > 1997-07-31 11:05:14 11670 11661 Begin backup logical log 4. > 1997-07-31 11:05:14 11670 11661 Successfully connected to Storage Manager. > > 1997-07-31 11:05:14 11670 11661 Completed backup logical log 4. > 1997-07-31 11:07:03 12094 12079 /vobs/tristarp/sqldist/bin/onbar -r -t > 1997\\170#^H-07\\170#^H-31\\170#^H 11\\170# > ^H:04\\170#^H:51 > > --Anil > > fitr0912@my-deja.com wrote: > > > Any ideas would be appreciated! > > > > I perform the following everyday at 22:00. > > > > onbar -b -L 0 -w> > > > It takes approx. 40 minutes to run. I got a request to perform a > > restore. I was to restore the instance to 22:38:22 the night before. I > > then issued the following command. > > > > onbar -r -t 1999-08-31 22:38:22> > > > The log number which contained this time was 5711. While monitoring the > > restore I noticed the following message in the online log. > > > > PIT reached - logid: 5711, logpos: 0x2e1158 > > > > I thought the restore would be completed soon. However, the restore > > continued for many hours. The restore was going through all the logs > > since 5711. I finally killed the restore. I then performed another > > restore. However, this time I chose to restore to the log number 5710. > > This worked perfectly. > > > > My environment is as follows: Informix 7.24.UCX1, HP Omniback II Version > > A.03.00, HP-UX 10.20 > > > > Thanks for any ideas! > > > > G > > > > Sent via Deja.com http://www.deja.com/ > > Share what you know. Learn what you don't. > > Sent via Deja.com http://www.deja.com/ Share what you know. Learn what you don't.