Sybase NNTP forums - End Of Life (EOL)

The NNTP forums from Sybase - forums.sybase.com - are now closed.

All new questions should be directed to the appropriate forum at the SAP Community Network (SCN).

Individual products have links to the respective forums on SCN, or you can go to SCN and search for your product in the search box (upper right corner) to find your specific developer center.

SFM cntrl hangs up

4 posts in General Discussion Last posting was on 2007-07-12 18:22:41.0Z
sbenn Posted on 2007-07-03 12:18:53.0Z
Sender: 4764.4672bf0c.1804289383@sybase.com
From: sbenn
Newsgroups: sybase.public.impact
Subject: SFM cntrl hangs up
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <468a3ead.48e5.1681692777@sybase.com>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 3 Jul 2007 05:18:53 -0700
X-Trace: forums-1-dub 1183465133 10.22.241.41 (3 Jul 2007 05:18:53 -0700)
X-Original-Trace: 3 Jul 2007 05:18:53 -0700, 10.22.241.41
Lines: 10
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.impact:2064
Article PK: 230146

Impact 5.4.6.8 on AIX5.2: we have seen twice now where a sfm
cntlr hangs up and doesnt respond to cmds from GC. The cmds
time out. GC shows the acquisition aims in failed mod (we
have source ping defined). Currently we bounce the cluster
to recover.
Has any one experienced this ?
Can I just kill -9 the sfm controller and then restart it
without bouncing the whole cluster?
thanks
steve


Christian Deschamps Posted on 2007-07-03 14:53:59.0Z
Sender: 72b0.4678177b.1804289383@sybase.com
From: Christian Deschamps
Newsgroups: sybase.public.impact
Subject: Re: SFM cntrl hangs up
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <468a6307.4d2e.1681692777@sybase.com>
References: <468a3ead.48e5.1681692777@sybase.com>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 3 Jul 2007 07:53:59 -0700
X-Trace: forums-1-dub 1183474439 10.22.241.41 (3 Jul 2007 07:53:59 -0700)
X-Original-Trace: 3 Jul 2007 07:53:59 -0700, 10.22.241.41
Lines: 18
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.impact:2065
Article PK: 230148

Can you check whether both applications (acq AIM and SFM)
try to reach each other mostly at the same time (the AIM
sends a call to the DFC route_v*** and SFM sends a call to
the DFC ping) ?

DFC deadlock?

> Impact 5.4.6.8 on AIX5.2: we have seen twice now where a
> sfm cntlr hangs up and doesnt respond to cmds from GC. The
> cmds time out. GC shows the acquisition aims in failed mod
> (we have source ping defined). Currently we bounce the
> cluster to recover.
> Has any one experienced this ?
> Can I just kill -9 the sfm controller and then restart it
> without bouncing the whole cluster?
> thanks
> steve


DOug Myers Posted on 2007-07-09 15:08:50.0Z
Sender: 4544.4676d557.1804289383@sybase.com
From: Doug Myers
Newsgroups: sybase.public.impact
Subject: Re: SFM cntrl hangs up
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <46924f82.1527.1681692777@sybase.com>
References: <468a6307.4d2e.1681692777@sybase.com>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 9 Jul 2007 08:08:50 -0700
X-Trace: forums-1-dub 1183993730 10.22.241.41 (9 Jul 2007 08:08:50 -0700)
X-Original-Trace: 9 Jul 2007 08:08:50 -0700, 10.22.241.41
Lines: 29
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.impact:2066
Article PK: 230149

I had this probelm for quite a while last year. This has
been seen by several other people also. The problem is AIX
5.2. You need to go to AIX 5.3 have maintenance level 3 or
higher, right now IBM has 5 out so I have used that. This
issue has to do with AIX TCP Keepalives which timeout and
cause controllers to hang and crash. Call Sybase support
they will be able to help you more specifically with your
issue.

> Can you check whether both applications (acq AIM and SFM)
> try to reach each other mostly at the same time (the AIM
> sends a call to the DFC route_v*** and SFM sends a call to
> the DFC ping) ?
>
> DFC deadlock?
>
>
> > Impact 5.4.6.8 on AIX5.2: we have seen twice now where a
> > sfm cntlr hangs up and doesnt respond to cmds from GC.
> > The cmds time out. GC shows the acquisition aims in
> > failed mod (we have source ping defined). Currently we
> > bounce the cluster to recover.
> > Has any one experienced this ?
> > Can I just kill -9 the sfm controller and then restart
> > it without bouncing the whole cluster?
> > thanks
> > steve


sbenn Posted on 2007-07-12 18:22:41.0Z
Sender: 4764.4672bf0c.1804289383@sybase.com
From: sbenn
Newsgroups: sybase.public.impact
Subject: Re: SFM cntrl hangs up
X-Mailer: WebNews to Mail Gateway v1.1t
Message-ID: <46967171.7a76.1681692777@sybase.com>
References: <468a3ead.48e5.1681692777@sybase.com>
NNTP-Posting-Host: 10.22.241.41
X-Original-NNTP-Posting-Host: 10.22.241.41
Date: 12 Jul 2007 11:22:41 -0700
X-Trace: forums-1-dub 1184264561 10.22.241.41 (12 Jul 2007 11:22:41 -0700)
X-Original-Trace: 12 Jul 2007 11:22:41 -0700, 10.22.241.41
Lines: 8
Path: forums-1-dub!not-for-mail
Xref: forums-1-dub sybase.public.impact:2069
Article PK: 230151

Even at AIX 5300-05 it happened. Checked the logs for dfc
deadlock and found it was hosing up on a transaction. lots
of testing it turns out its a production object. the po
worked fine in 4.1 but when ported to 5x under certain
situations would cause sfm to hose up. Unfortunately its an
ODl function with about 300 lines of code in it...
thanks for the help.
> steve