Highlighted

The suddenly and inexplicably RESTART status with all the TDs of some media servers!

Sometimes running oprcmd -devmon ds would output the RESTART status with all the TDs of some media servers while they were all UP status via tpconfig -l on there own media  servers!
Notes: in most or normal case, the TDs status of oprcmd -devmon ds and the corresponding tpconfig -l should match!
But here the media servers' TDs status of oprcmd -devmon ds would suddenly and inexplicably became RESTART!
And at last we have to manually restart the ltid process for the corresponding media servers,then all recovered finely!

root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -devmon ds|grep -Ei "avr|down|pen|rest"|wc -l
0
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -devmon ds|grep -Ei "avr|down|pen|rest"|wc -l
22
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -devmon ds|grep -Ei "avr|down|pen|rest"
gwzycdb5 /dev/nst5 RESTART
gwzycdb2 /dev/nst3 RESTART
gwzycdb3 /dev/nst7 RESTART
gwzycdb5 /dev/nst7 RESTART
gwzycdb2 /dev/nst2 RESTART
gwzycdb3 /dev/nst6 RESTART
gwzycdb5 /dev/nst6 RESTART
gwzycdb2 /dev/nst0 RESTART
gwzycdb4 /dev/nst4 RESTART
gwzycdb5 /dev/nst4 RESTART
gwzycdb2 /dev/nst4 RESTART
gwzycdb3 /dev/nst3 RESTART
gwzycdb4 /dev/nst3 RESTART
gwzycdb2 /dev/nst5 RESTART
gwzycdb3 /dev/nst2 RESTART
gwzycdb4 /dev/nst2 RESTART
gwzycdb2 /dev/nst7 RESTART
gwzycdb3 /dev/nst1 RESTART
gwzycdb4 /dev/nst1 RESTART
gwzycdb2 /dev/nst6 RESTART
gwzycdb3 /dev/nst0 RESTART
gwzycdb4 /dev/nst0 RESTART
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -devmon ds|grep -Ei "avr|down|pen|rest"|wc -l
22
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -devmon ds|grep -Ei "avr|down|pen|rest"|grep gwzycdb3
gwzycdb3 /dev/nst7 RESTART
gwzycdb3 /dev/nst6 RESTART
gwzycdb3 /dev/nst3 RESTART
gwzycdb3 /dev/nst2 RESTART
gwzycdb3 /dev/nst1 RESTART
gwzycdb3 /dev/nst0 RESTART
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#

 

root@gwzycdb3:/root#
root@gwzycdb3:/root#
root@gwzycdb3:/root# tpconfig -l
Device Robot Drive Robot Drive Device Second
Type Num Index Type DrNum Status Comment Name Path Device Path
robot 1 - TLD - - - - jcbak
drive - 0 hcart3 6 UP - HP.ULTRIUM6-SCSI.004 /dev/nst3
drive - 1 hcart3 5 UP - HP.ULTRIUM6-SCSI.005 /dev/nst2
drive - 2 hcart3 7 UP - HP.ULTRIUM6-SCSI.006 /dev/nst1
drive - 3 hcart3 8 UP - HP.ULTRIUM6-SCSI.007 /dev/nst0
drive - 4 hcart3 2 UP - HP.ULTRIUM6-SCSI.001 /dev/nst7
drive - 5 hcart3 1 UP - HP.ULTRIUM6-SCSI.002 /dev/nst6
root@gwzycdb3:/root#
root@gwzycdb3:/root#


root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# date
Mon Nov 4 15:30:57 CST 2019
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -h gwzycdb3 -stopltid
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -h gwzycdb3 -startltid -v
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -devmon ds|grep -Ei "avr|down|pen|rest"|grep gwzycdb3|wc -l
0
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#
root@jcbak:/# vmoprcmd -devmon ds|grep -Ei "avr|down|pen|rest"|grep gwzycdb3
root@jcbak:/#
root@jcbak:/#
root@jcbak:/#

 

Tags (2)
1 Reply

Re: The suddenly and inexplicably RESTART status with all the TDs of some media servers!

@liuyl 

Who are you shouting at with all the exlamation marks ? !!!

If you need to know why the drives go into RESTART mode, you will need to dig into volmgr logs (if enabled) on the media servers to see why NBU believes that changes have been made to the device configuration that requires LTID to be restarted.