onunload/onload problem using tape

HPUX B.11.31 U ia64 on HP BL870c IDS 11.50.FC6 We have two HP BL870c Integrity servers, one is the production the second is for testing and recovery if production server fails. The systems have been in place for about 6 years and share two lto2 tape drive via fiber. Today I was trying to move a test DB from the test server to production server using onunload (/opt/informix/bin/onunload -t /dev/rmt/c16t0d1BEST cars). Have done this numerous times in the past. It has been about 6 month since the developers needed me to do this. I know of no changes to the hardware, OS or tape drives. The onumload command seemed to have worked, but when I try to do the onload command on the production server (/opt/informix/bin/onload -t /dev/rmt/c16t0d1BEST -d dbs1 train), I'm getting the following Mismatch in tape block size -- tape block size = 32K. d press Return to continue ... I have a support call in to HP to see what's going on, but so far has not been helpful. My 10+ year old training on onload/unload is rusty and I am thinking the block size written to the tape is determined by the OS/tape drive and not Informix. Is this correct? I want to make sure this is a hardware/OS issue and really has nothing to do with Informix - other than is causing the onload to fail. John David Adamski, Sr. Network Specialist Graceland University, 1 University Place, Lamoni, IA 50140

On Sunday, June 24, 2012 12:48:35 PM UTC-5, jda wrote: > HPUX B.11.31 U ia64 on HP BL870c > IDS 11.50.FC6 > > We have two HP BL870c Integrity servers, one is the production the second is for testing and recovery if production server fails. The systems have been in place for about 6 years and share two lto2 tape drive via fiber. > > Today I was trying to move a test DB from the test server to production server using onunload (/opt/informix/bin/onunload -t /dev/rmt/c16t0d1BEST cars). Have done this numerous times in the past. It has been about 6 month since the developers needed me to do this. I know of no changes to the hardware, OS or tape drives. > > The onumload command seemed to have worked, but when I try to do the onload command on the production server (/opt/informix/bin/onload -t /dev/rmt/c16t0d1BEST -d dbs1 train), I'm getting the following > > Mismatch in tape block size -- tape block size = 32K. > d press Return to continue ... > > I have a support call in to HP to see what's going on, but so far has not been helpful. > > My 10+ year old training on onload/unload is rusty and I am thinking the block size written to the tape is determined by the OS/tape drive and not Informix. Is this correct? I want to make sure this is a hardware/OS issue and really has nothing to do with Informix - other than is causing the onload to fail. > > John David Adamski, Sr. Network Specialist > Graceland University, 1 University Place, Lamoni, IA 50140 Sorry in advance if this double posts. Anyway, a quick look in the source and it shows that onunload/onload will pick up a default block size based on the value you have for TAPEBLK in your $ONCONFIG file. My guess would be that TAPEBLK is no longer the same between your two instances, so you now have to use the -b flag to over ride the default getting picked up from $ONCONFIG. Jacques Renaut IBM Informix Advanced Support APD Team

Jacques Yep, somehow the product was set to 64 and the test server is set to 32. Hmmmm, someone been playing with things they are not supposed to change. Guess I need to see what else got changes. Thanks for your reply John -----Original Message----- From: informix-list-bounces@iiug.org [mailto:informix-list-bounces@iiug.org] On Behalf Of jrenaut Sent: Monday, June 25, 2012 10:56 AM To: informix-list@iiug.org Cc: informix-list@iiug.org Subject: Re: onunload/onload problem using tape On Sunday, June 24, 2012 12:48:35 PM UTC-5, jda wrote: > HPUX B.11.31 U ia64 on HP BL870c > IDS 11.50.FC6 > > We have two HP BL870c Integrity servers, one is the production the second is for testing and recovery if production server fails. The systems have been in place for about 6 years and share two lto2 tape drive via fiber. > > Today I was trying to move a test DB from the test server to production server using onunload (/opt/informix/bin/onunload -t /dev/rmt/c16t0d1BEST cars). Have done this numerous times in the past. It has been about 6 month since the developers needed me to do this. I know of no changes to the hardware, OS or tape drives. > > The onumload command seemed to have worked, but when I try to do the > onload command on the production server (/opt/informix/bin/onload -t > /dev/rmt/c16t0d1BEST -d dbs1 train), I'm getting the following > > Mismatch in tape block size -- tape block size = 32K. > d press Return to continue ... > > I have a support call in to HP to see what's going on, but so far has not been helpful. > > My 10+ year old training on onload/unload is rusty and I am thinking the block size written to the tape is determined by the OS/tape drive and not Informix. Is this correct? I want to make sure this is a hardware/OS issue and really has nothing to do with Informix - other than is causing the onload to fail. > > John David Adamski, Sr. Network Specialist Graceland University, 1 > University Place, Lamoni, IA 50140 Sorry in advance if this double posts. Anyway, a quick look in the source and it shows that onunload/onload will pick up a default block size based on the value you have for TAPEBLK in your $ONCONFIG file. My guess would be that TAPEBLK is no longer the same between your two instances, so you now have to use the -b flag to over ride the default getting picked up from $ONCONFIG. Jacques Renaut IBM Informix Advanced Support APD Team _______________________________________________ Informix-list mailing list Informix-list@iiug.org http://www.iiug.org/mailman/listinfo/informix-list