[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference vmsdev::vmstuning

Title:VMSTUNING
Notice:Welcome to VMSTUNING
Moderator:EVMS::HALLYB
Created:Sat Feb 15 1986
Last Modified:Wed May 14 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:1539
Total number of notes:7984

1532.0. "$stop/id command does not delete." by DEKVC::CHANGHOANN () Thu Feb 27 1997 20:08

Hello,

One of our big customers has a process hang problem.
This process is Oracle RDB application process, stalls in the state of 
exclusive mode lock for RDB database file. Because all rdb processes accessing 
the RDB file are blocked by this stall process, the customer tried the DCL 
command "STOP/IDENTIFICATION". But the process is not deleted, goes into
SUSP state. After all, the customer rebooted the system.

Question:

Is this problem related to VMS side ? If not, should ORACLE RDB take care this 
hang problem ?

Is there any workaround to delete the stalled process ?

Thanks in advance for any recommendation.

Chang-ho Ann/
T.RTitleUserPersonal
Name		DateLines
1532.1probably resource wait of some sortGIDDAY::GILLINGSa crucible of informative mistakesThu Feb 27 1997 23:0220
  Chang-ho Ann,

> But the process is not deleted, goes into
> SUSP state. After all, the customer rebooted the system.

    I'll bet it's NOT in SUSP state at all. More likely an RWAST of some
  description. In order to diagnose this problem, it is vital that 
  someone who knows what they're doing has a look at this process BEFORE
  anyone attempts to STOP/ID it. 

> Is this problem related to VMS side ?

    Could be any of a very large number of causes. Knowing the O/S version
  and what patches are installed might help.

> Is there any workaround to delete the stalled process ?

    Maybe and maybe not. The process would need to be examined BEFORE the
  STOP to say for sure.
						John Gillings, Sydney CSC
1532.2DEKVC::CHANGHOANNSun Mar 02 1997 19:3332
John Gillings,

Thanks for your reply.

>  I'll bet it's NOT in SUSP state at all. More likely an RWAST of some
>  description. 
>  In order to diagnose this problem, it is vital that 
>  someone who knows what they're doing has a look at this process BEFORE
>  anyone attempts to STOP/ID it. 

I agree with you. If it is in RWAST state, and there is not any h/w fault, 
please let me know what I should do to check. I think there is a stalled RDB 
database lock. I don't know RDB.

>> Is this problem related to VMS side ?
>    Could be any of a very large number of causes. Knowing the O/S version
>  and what patches are installed might help.

O/s version : OpenVMS/Alpha V6.2-1H3
	      Alpha 7740*3 CI cluster
The installed patches:
	ALPDRIV03_070,ALPDRIV04_062,ALPDRIV05_062,ALPF1102_070, ALPLOGI02_070
	ALPMANA01_070,ALPMSCP01_062,ALPRMS01_062, ALPSCSI02_070,ALPSYS04_062
	ALPSYS05_070, ALPSHAD05_062,ALPBOOT05_062,ALPCLIU01_062,ALPDDTM03_070
	ALPDISM01_070,ALPF1103_070, ALPLAD03_070, ALPLIBR04_070,ALPLOAD02_070
	ALPMTAA01_070,ALPOPC01_070, ALPVERI01_062

Do you have any other idea ?

Regards,
Chang-ho Ann