alphastor with two tape libraries 3494 on the save backup server fails

Can you mount/unmount tapes using AS CLI? Do the tapes get mounted to the drive you would expect so?

If yes, try then to run nsrjb command from bck server in verbose mode and see what happens. I guess this is very it breaks. Can you communicate at all with AS using existing NW jukebox definition - try to run inventory from NW.

When doing upgrade did you reconfigure libraries (configure them again) in AS? If yes, I would suggest to do the same on NW side.

The problem at the moment does not involve NetWorker. The issue is between alphastor and the tap libaries.

Now we are trying to come back to the initial release of the software but alphastor seems to be unable to unmount tape. Mount operations are ok. During the unmount instead the process hungs

follows the truss of the unmount process.

recv(3, 0x00058C20, 10240, 0) (sleeping...)
signotifywait() (sleeping...)
door_return(0x00000000, 0, 0x00000000, 0) (sleeping...)
lwp_sema_wait(0xFEE0BE60) (sleeping...)
lwp_cond_wait(0xFF1334E8, 0xFF1334F8, 0xFF12CD80) (sleeping...)
lwp_cond_wait(0xFF1334E8, 0xFF1334F8, 0xFF12CD80) Err#62 ETIME
recv(3, 0x00058C20, 10240, 0) (sleeping...)
signotifywait() (sleeping...)
door_return(0x00000000, 0, 0x00000000, 0) (sleeping...)
lwp_sema_wait(0xFEE0BE60) (sleeping...)
lwp_cond_wait(0xFF1334E8, 0xFF1334F8, 0xFF12CD80) (sleeping...)
recv(3, "\0\0\004\0\0\001", 10240, 0) = 8
sigaction(SIGPIPE, 0xFFBEF650, 0x00000000) = 0
shutdown(3, 2, 1) = 0
close(3) = 0
write(2, " U n m o u n t f a i l".., 16) = 16
write(2, " 1 )\n", 3) = 3
llseek(0, 0, SEEK_CUR) = 122029
_exit(2)

Yes, but at the moment you don't have any activity and only activity you will have is asdismount which we expect to fail, right? Giving that to support may trigger solution or understanding what is going on much faster.

You mean, enabling debugging by /etc/asenv? Because we know that this operation enable the debug for every AS activities and log file grew too mutch.

Enable debugging on AS and try to mount/dismount using asvolmount/asvoldismount. When it fails get the debug output and see if the error does any sense. If not open the case with support. They will probably ask about the steps you did during the unpgrade so make sure to have those ready. I don't think that going back to initial release will help and something basic seems to be wrong - easiest thing is to suspect that during upgrade process something was changed and messed current state (if so going back won't help).

Ok, we try are trying just now.
Please, could you suggest me the exact procedure to activate the AS debugging.
It is the first time I have to do it.

I don't know it now top of my head, but if I remember correctly it should say what is needed in comments - it's pretty straight process. Check also if you have AS_HOST listed there - previous version had some issues. You can copy and paste your /etc/asenv here and we can try to modify it. Otherwise appendix B of Ops guide has the details.

Mmm, wait. LGTpa61890 is not fixed by 3.0.3 but rather by SR2 meaning this could be your issue (symptoms do match). If you needed that patch before you will need it now with 3.0.3 too - open this with support.

/etc/asenv.

AS_CLIDPORT 41025
AS_ASLOGPORT 41114 # Date=04-05-2004
AS_ASQMPORT 44460 # Date=04-05-2004
AS_ASMTPORT 44455 # Date=04-05-2004
AS_UPDATENW ON # Date=04-05-2004
AS_VOLLIST ON # Date=04-05-2004
RESOURCE_QUEUE_TM_ENABLE ON # Date=13-04-2004
AS_HOST alpstsrv # AlphaStor Host Date=04-05-2004
ASHOME /usr/local/as # AlphaStor home directory Date=04-05-2004
ROB_HOST alpstsrv # host name for this system Date=04-05-2004
ROB_HOME /usr/local/as # robot home path for this system Date=04-05-2004
ROB_REMOTE_ROBOT alpstsrv:robotd # default server host
ROB_REMOTE_ROBOT_LIB3494A alpstsrv:robotd # server host for LIB3494A
ROB_REMOTE_ROBOT_LIB3494U alpstsrv:robotd # server host for LIB3494U




At the moment we have tried to dismount tape by nsrjb and this is what happened


/tmp/panic.log:

Wed Mar 1 15:11:08: [1] [11] Begin move tape from drive vid UX0229 device D7926 Ver @(#) AlphaStor 3.0.3_33
Wed Mar 1 15:11:08: [1] [11] Tape drive D7926 identified (d79260)
Wed Mar 1 15:11:08: [1] [11] Dismounting tape UX0229 device D7926 address d79260
Wed Mar 1 15:11:08: [1] [11] ERROR: Could not remove tape from D7926 Version: @(#) AlphaStor 3.0.3_33

Wed Mar 1 15:11:08: [13] Could not eject tape nwservbe drive D7926 all hosts tried
Wed Mar 1 15:11:08: [11] read response complete status 0 data
Wed Mar 1 15:11:08: [11] read response complete status 13 data 0
Wed Mar 1 15:11:19: [11] read response complete status 0 data 2 -1 1

nsrjb -uv:

Couldnt unmount tape=
Error class: AS UNMOUNT, Error token: UNLOADFAIL, Error text: Couldnt unmount tape.
slot number 30, SmartMedia error class: 'AS UNMOUNT', token: 'UNLOADFAIL', Couldnt unmount tape
failed unloading drive `U2' to slot '30', error '0'

Error: while operating on slot `30': SmartMedia error class: 'AS UNMOUNT', token: 'UNLOADFAIL', Couldnt unmount tape
Wed 15:11:09 SmartMedia error class: 'AS UNMOUNT', token: 'UNLOADFAIL', Couldnt unmount tape

nsrjb: Jukebox error: Wed 15:11:09 SmartMedia error class: 'AS UNMOUNT', token: 'UNLOADFAIL', Couldnt unmount tape



We have installed and restored the original configuration, db too, that is:
alphastor 3.0 SR1
with the following patches: LGTpa62746, LGTpa61890

If we have completely restored the previews configuration what should be changed?

ok, thks for your support.

And it was dor 3.0.1 as you said. And according to my notes this LGTpa is not fixed in 3.0.3 so if you had an issue then then you would have an issue now as well. Only SR2 did address this issue. So, it is hard to say what it is now - that issue or something else or mix. That's why your best bet now are official case and official statements from official ppl.

We find out the cause of the issue. The default alphastor communication port was changed during the previous installation.

Great! Now ask them also for the patch - if you did use it before you will still need it with 3.0.3.