home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!mips!swrinde!elroy.jpl.nasa.gov!decwrl!decwrl!infopiz!mccall!ipmdf-newsgate!list
- From: dan@innosoft.com (Daniel C. Newman)
- Newsgroups: vmsnet.mail.pmdf
- Subject: RE: pmdf v4.1-1 and slow delivery
- Message-ID: <01GNX7ILY3TE94E1EZ@INNOSOFT.COM>
- Date: 23 Aug 92 21:12:31 GMT
- Organization: The Internet
- Lines: 29
- Return-Path: <epmdf@YMIR.CLAREMONT.EDU>
- Resent-Date: 23 Aug 1992 13:12:31 -0800 (PST)
- Resent-From: epmdf@YMIR.CLAREMONT.EDU
- Errors-To: epmdf@YMIR.CLAREMONT.EDU
- Resent-Message-ID: <01GNX7JJM5QA95O1A9@YMIR.CLAREMONT.EDU>
- X-Vms-To: IN%"tkslen@ubvms.cc.buffalo.edu"
- X-Vms-Cc: IPMDF
- Mime-Version: 1.0
- Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
- Content-Transfer-Encoding: 7BIT
-
- > I just installed vms v5.5-1 and pmdf v4.1-1 on monday. Ever since then it
- > seems
- > my queues build during the day (all of them from the local to the bitnet and
- > internet queues) then during the night they catch up. I feel this is from the
- > low volume of new mail during the night.
- > Anyone have any idea's of what could be causing the performance problem. The
- > cpu the queue is running on not 100% loaded. The log files from the delivery's
- > show no errors, just successful delivery's.
- > When I installed pmdf all I did was run bitnet_domains. Should I have run
- > something else?
-
- This sounds a lot like a queue cache ownership problem. If the ownership ot
- the queue cache is incorrect, then some incoming mail (depends upon the
- channel) may not get registered in the queue cache and thus will sit around
- undelivered until it is registered in the queue cache. At about midnight, the
- RETURN job will run and will synchronize the cache and finally some of this
- mail will be noticed and delivered.
-
- The queue cache is the file PMDF_ROOT:[TABLE]QUEUE_CACHE.DAT. It should be
- owned by [PMDF] or whatever your PMDF account is. If you do not have a PMDF
- account, then it should be owned by the [SYSTEM]. Its protection should be
- (S:RWED,O:RWED,G,W). There was a problem with 4.1-0, -1, and -2 kits which
- could lead to the queue cache not having the proper ownership.
-
- ... and then your problem may be something entirely different. If that is the
- case, the best way to start in tracking the problem down is to identify which
- channels seem to be afflicted the most.
-
- Dan
-