Backup hanging

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Dirk Moolman - 29 Apr 2005 08:36 GMT

I have a backup that is just hanging - tape drive is idle, reads and
writes are not going up:

3431c5ce8 ------- 5644203 root - 0 0
0 137996888 0
345c698e8 ------- 5644203 root - 0 0
0 0 119790
364e33248 Y-AP--M 5644203 root - 342420a38 0
0 3386 40

Onstat -g ses shows that it is waiting for something:

Informix Dynamic Server Version 7.31.FD3 -- On-Line -- Up 41 days
01:13:16 -- 11826176 Kbytes

session #RSAM total used
id user tty pid hostname threads memory memory
5644203 root - 26211 mxsun 3 753664 623744

tid name rstcb flags curstk status
5674033 ontape 364e33248 Y-AP--M 4879 cond wait(netnorm)
5674036 arcbacku 3431c5ce8 ------- 3791 sleeping(secs: 1)
5674037 arcbacku 345c698e8 ------- 2879 sleeping(secs: 1)

And Unix also shows that it is sleeping:

truss -p 26211
*** SGID: rgid/egid/sgid = 1 / 200 / 200 ***
read(0, 0x10033F714, 5120) (sleeping...)

Any ideas ? My logs are not full, Informix is operational (no
blocked checkpoints or long transactions)

Dirk

sending to informix-list

Reply to this Message

Paul Watson - 29 Apr 2005 11:48 GMT

Are you running from a script? Netnorm normally equates back to 'press
any key to continue'

> I have a backup that is just hanging - tape drive is idle, reads and
> writes are not going up:
[quoted text clipped - 32 lines]
>
> sending to informix-list

Signature

Paul Watson #
Oninit Ltd # Growing old is mandatory
Tel: +44 1436 672201 # Growing up is optional
Fax: +44 1436 678693 #
Mob: +44 7818 003457 #
www.oninit.com #

Reply to this Message

Dirk Moolman - 29 Apr 2005 13:59 GMT

Yes, I am running from a script, but the script was written to pass all
the necessary parameters to Informix (and has been running successfully
for months now).

We managed to trace the problem, using "truss" and "lsof" (which you
have to download and install), to debug the Unix process.

What apparently happened, was that the script pipes parameters to
ontape. The network dropped. The script was not aware of the drop, and
sat waiting for input from the network (condition wait:netnorm, and unix
process sleeping)

Guess the network admin has some troubleshooting to do .....

Thanks for the help.
Dirk

-----Original Message-----
From: owner-informix-list@iiug.org [mailto:owner-informix-list@iiug.org]
On Behalf Of Paul Watson
Sent: 29 April 2005 12:48 PM
To: informix-list@iiug.org
Subject: Re: Backup hanging

Are you running from a script? Netnorm normally equates back to 'press
any key to continue'

> I have a backup that is just hanging - tape drive is idle, reads and
> writes are not going up:
[quoted text clipped - 32 lines]
>
> sending to informix-list

Signature

Paul Watson #
Oninit Ltd # Growing old is mandatory
Tel: +44 1436 672201 # Growing up is optional
Fax: +44 1436 678693 #
Mob: +44 7818 003457 #
www.oninit.com #

sending to informix-list

Reply to this Message