Ontape Restore

I am trying to restore a dbspace from one instance to another instance on another server. The instances are setup exactly the same except for mirroring -- the dbspaces have the same names, the chunks have the same names, sizes and offsets, and the logical and physical logs are the same size and number. Each time I try to complete the restore, I get an error that says 'Physical restore failed - Bad log file configuration for log 1.' I have double checked the logical and physical logs to make sure they are them same, but I can't find anything wrong or different. Any clues? I am running Informix 7.30.UC5 on HP-UX 10.2. thanks.

Some questions that may help broaden the understanding of your problem :- Is your "mirroring" Informix mirroring or hardware/os level mirroring? What commands did you run for the restore? Diff the onconfig files from the instances. Complete the restore? At what stage did the error occur? What is the status of the target instance? Can it be initialized? Brett Laurie Gustin wrote: > I am trying to restore a dbspace from one instance to another instance on > another server. The instances are setup exactly the same except for > mirroring -- the dbspaces have the same names, the chunks have the same > names, sizes and offsets, and the logical and physical logs are the same > size and number. > > Each time I try to complete the restore, I get an error that says 'Physical > restore failed - Bad log file configuration for log 1.' > > I have double checked the logical and physical logs to make sure they are > them same, but I can't find anything wrong or different. Any clues? > > I am running Informix 7.30.UC5 on HP-UX 10.2. > > thanks.

the mirroring is on the Informix level the command was ontape -r -D dbspacename the onconfig files are the same, except for mirring the rootdbspace the restore failed after about 15 minutes, the log file showed the restore begin, but no data was written ( as per onstat -u ) the target instance is initialized, the target dbspace is down, the rest of the instance is empty with dbspaces created, but no data. "Brett Randall" <Brett@reply.to.newsgroup> wrote in message news:3A01509B.F057C9BD@reply.to.newsgroup... > Some questions that may help broaden the understanding of your problem :- > > Is your "mirroring" Informix mirroring or hardware/os level mirroring? > What commands did you run for the restore? > Diff the onconfig files from the instances. > Complete the restore? At what stage did the error occur? > What is the status of the target instance? Can it be initialized? > > Brett > > Laurie Gustin wrote: > > > I am trying to restore a dbspace from one instance to another instance on > > another server. The instances are setup exactly the same except for > > mirroring -- the dbspaces have the same names, the chunks have the same > > names, sizes and offsets, and the logical and physical logs are the same > > size and number. > > > > Each time I try to complete the restore, I get an error that says 'Physical > > restore failed - Bad log file configuration for log 1.' > > > > I have double checked the logical and physical logs to make sure they are > > them same, but I can't find anything wrong or different. Any clues? > > > > I am running Informix 7.30.UC5 on HP-UX 10.2. > > > > thanks. >

I'm not sure that the operation you are trying to perform is valid ... I think of database restores as "recovering consistency" rather than "restoring the lost data/disk". In your example, I assume that you started with all dbspaces consistent in both instances A and B. Then you "trashed" a dbspace in instance B and are attempting a recovery from a backup of that dbspace from in instance A. However, this would only make the dbspace consistent with the other dbspaces in instance A, not the dbspaces in instance B. The only way to do what you want is to start with Instance B in offline mode, do a cold restore including all critical dbspaces (root, logs), and the dbspace you want restored. Having said all this, your error message also appears a lot in the Informix Tech Info database. HTH Brett Randall Laurie Gustin wrote: > the mirroring is on the Informix level > the command was ontape -r -D dbspacename > the onconfig files are the same, except for mirring the rootdbspace > the restore failed after about 15 minutes, the log file showed the restore > begin, but no data was written ( as per onstat -u ) > the target instance is initialized, the target dbspace is down, the rest of > the instance is empty with dbspaces created, but no data. > > "Brett Randall" <Brett@reply.to.newsgroup> wrote in message > news:3A01509B.F057C9BD@reply.to.newsgroup... > > Some questions that may help broaden the understanding of your problem :- > > > > Is your "mirroring" Informix mirroring or hardware/os level mirroring? > > What commands did you run for the restore? > > Diff the onconfig files from the instances. > > Complete the restore? At what stage did the error occur? > > What is the status of the target instance? Can it be initialized? > > > > Brett > > > > Laurie Gustin wrote: > > > > > I am trying to restore a dbspace from one instance to another instance > on > > > another server. The instances are setup exactly the same except for > > > mirroring -- the dbspaces have the same names, the chunks have the same > > > names, sizes and offsets, and the logical and physical logs are the same > > > size and number. > > > > > > Each time I try to complete the restore, I get an error that says > 'Physical > > > restore failed - Bad log file configuration for log 1.' > > > > > > I have double checked the logical and physical logs to make sure they > are > > > them same, but I can't find anything wrong or different. Any clues? > > > > > > I am running Informix 7.30.UC5 on HP-UX 10.2. > > > > > > thanks. > >

If you use Informix mirroring you have to have identical files available for the mirror chunks as well as the primary chunks. Art S. Kagel Laurie Gustin wrote: > > the mirroring is on the Informix level > the command was ontape -r -D dbspacename > the onconfig files are the same, except for mirring the rootdbspace > the restore failed after about 15 minutes, the log file showed the restore > begin, but no data was written ( as per onstat -u ) > the target instance is initialized, the target dbspace is down, the rest of > the instance is empty with dbspaces created, but no data. > > "Brett Randall" <Brett@reply.to.newsgroup> wrote in message > news:3A01509B.F057C9BD@reply.to.newsgroup... > > Some questions that may help broaden the understanding of your problem :- > > > > Is your "mirroring" Informix mirroring or hardware/os level mirroring? > > What commands did you run for the restore? > > Diff the onconfig files from the instances. > > Complete the restore? At what stage did the error occur? > > What is the status of the target instance? Can it be initialized? > > > > Brett > > > > Laurie Gustin wrote: > > > > > I am trying to restore a dbspace from one instance to another instance > on > > > another server. The instances are setup exactly the same except for > > > mirroring -- the dbspaces have the same names, the chunks have the same > > > names, sizes and offsets, and the logical and physical logs are the same > > > size and number. > > > > > > Each time I try to complete the restore, I get an error that says > 'Physical > > > restore failed - Bad log file configuration for log 1.' > > > > > > I have double checked the logical and physical logs to make sure they > are > > > them same, but I can't find anything wrong or different. Any clues? > > > > > > I am running Informix 7.30.UC5 on HP-UX 10.2. > > > > > > thanks. > >