ADSM-L

adsm server hangs- update 1

2001-01-23 04:51:10
Subject: adsm server hangs- update 1
From: Heinrich Hartl <heinrich.hartl AT POWER.ALSTOM DOT COM>
Date: Tue, 23 Jan 2001 10:52:05 +0100
Thank you Fred and Jack for first responses.

With your encouragement in mind I tried to halt the server.
adsm> disable sessions
adsm> halt
However the server just doesn't halt. As documented it seems to try to cancel
the existing sessions (which is not succesful explicitly) and doesn't stop
before this is succesful. In the meantime a few more client session have been
opened this night.
I also tried to cancel the client sessions from the client side.
e.g.
mrclx #> cancel <dsmsched-pid>
The client processes were successfully stopped. However the server sessions as
seen by
adsm> q sess
did not terminate.

Though halt has been issued I stll can connect to the server with a dsmadmc
session.

Regards
Heinrich Hartl
============================================= initial call for help
I've run into a problem and I would be very obliged if someone has an idea what
I could  do about it:

ADSM Server for AIX-RS/6000 - Version 3, Release 1, Level 2.20

seems to be in a strange state and nomore doing what its expected to do.

dsmadmc shows the following informations:

adsm> q log
Available Assigned   Maximum   Maximum    Page     Total      Used   Pct  Max.
    Space Capacity Extension Reduction    Size    Usable     Pages  Util   Pct
     (MB)     (MB)      (MB)      (MB) (bytes)     Pages                  Util
--------- -------- --------- --------- ------- --------- --------- ----- -----
      248      248         0       240   4.096    62.976       773   1,0  10,0
      248      248         0       240   4.096    62.976       773   1,0  10,0

adsm> q db
Available Assigned   Maximum   Maximum    Page     Total      Used   Pct  Max.
    Space Capacity Extension Reduction    Size    Usable     Pages  Util   Pct
     (MB)     (MB)      (MB)      (MB) (bytes)     Pages                  Util
--------- -------- --------- --------- ------- --------- --------- ----- -----
    2.288    1.476       812       244   4.096   377.856   247.144  65,0  79,0
    2.288    1.476       812       244   4.096   377.856   247.144  65,0  79,0

sofar anything  normal.
====================================

adsm> q sess
  Sess Comm.  Sess     Wait   Bytes   Bytes Sess  Platform Client Name
Number Method State    Time    Sent   Recvd Type
------ ------ ------ ------ ------- ------- ----- -------- --------------------
 2.072 Tcp/Ip Run      0 S   88,0 M 339,0 M Node  WinNT    ACLUSTER
 2.072 Tcp/Ip Run      0 S   88,0 M 339,0 M Node  WinNT    ACLUSTER
 2.119 Tcp/Ip Run      0 S      124     131 Admin AIX      ADMIN
 2.120 Tcp/Ip Run      0 S    5,0 K     359 Admin AIX      ADMIN
 2.121 Tcp/Ip Run      0 S    9,0 K     462 Admin AIX      ADMIN
 2.122 Tcp/Ip Run      0 S    9,0 K     657 Admin AIX      ADMIN
 2.126 Tcp/Ip Run      0 S  286,0 K     511 Admin AIX      ADMIN

## sessions 2072, 2119...2122 have been cancelled however they continue to hang
around
## The sessions arose from dsmadmc commands that would not finish and were
terminated with ^c
## such commands are
## q stg
## q vol

adsm> q pr
 Process Process Description  Status
  Number
-------- -------------------- -------------------------------------------------
      95 DELETE FILESPACE     Cancel in Progress.
      95 DELETE FILESPACE     Cancel in Progress.
      96 Database Backup      Cancel pending.
## Both processes were still executing showing no errors when the problem first
came up
## process status showed around 115k files deleted (prior to my cancel pr 95
request)
## process 96 was an incremental backup of the database with the last message in
the actlog that the output volume was dismounted

Trying to solve the problem (q vol, q stg no more responding) I subsequently
cancelled processes and sessions.
However processes did not end, sessions did not terminate.

I have not yet tried to halt the server.
(I am afraid the halt might hang as well since the processes/sessions do not
terminate)
A suspected reason for the problem is that a high volume of incremenatal backup
data may have been coming in while the delete filespace for a different node was
running.
<Prev in Thread] Current Thread [Next in Thread>
  • adsm server hangs- update 1, Heinrich Hartl <=