Have checked as suggested with monitor utility: The pauses are only 3-4
seconds where comm threads are climbing up to about 70 (this happens about
8-10 times every 15 minutes). The peak value for comm threads is 125 (max.
128) so it looks like that there are phases where the pauses might be
longer. Is there a chance to check how often the system is reaching the
peak
value without looking on the monitor for hours? If the system reaches the
peak value very often it will might worthwhile to monitor it for getting a
dump.
Of course you are right - advanced output was not enabled. Have checked
again but could not see a lot more paging writes during the pauses.
Michael
"Bill Bach" <goldstar@[EMAIL PROTECTED]
> wrote in message
news:mpednaFRMussORvanZ2dnUVZ_v-hnZ2d@[EMAIL PROTECTED]
> Bad timing? Perhaps. Might be worth trying again. See my reply to
> Brad for a suggestion on how to trap it a bit easier.
>
> How long are the pauses in your environment? I have seen some
> customers complaining about 3-5 MINUTES at times, but after switching
> to a RAID10 volume, they dropped to only seconds.
>
> As the problem has been quite evident with PSQLv9.5, I doubt that XIO
> has anything to do with it, so this info doesn't surprise me. However,
> it is odd that you are not seeing paging. Did you remember to enable
> "Advanced" output in FileMon? If you don't, you don't get the server's
> accesses, including the pagefile.
> Goldstar Software Inc.
> Pervasive-based Products, Training & Services
> Bill Bach
> BillBach@[EMAIL PROTECTED]
> http://www.goldstarsoftware.com
> *** Chicago: Pervasive Service & Sup****t Class - March 2008 ***
>
>
> michael last wrote:
>
>> Hi Bill,
>>
>> thanks very much for your info. "Good" to know that we are not the
>> only one with that problem.
>>
>> Today we have checked disk writes to pagefile with Filemon - NO disk
>> writes to pagefile, also nothing during high cpu utilization on
>> ntdbsmgr.
>>
>> We have already opened an incident directly with pervasive and sent
>> them a user dump. The analyze result from Pervasive: "The dump only
>> showed three active threads, and one of those was getting an MKDE
>> Stat on a file. There was no IO going on either, which led him to
>> believe that the dump was taken just after the pause cleared up."
>>
>> I think getting a dump containing these "pause" states is moreless
>> impossible since the pauses are not long enough.
>>
>> Just to let you know - we have also disabled new XIO feature in PSQL
>> 10 without any changes.
>>
>> Could the problem be a network communication issue? Faulty network
>> cards, routers, switches or outdated network drivers?
>>
>> Any help is highly appreciated since we have to update a lot more
>> clients to PSQL 10.
>>
>> Michael
>
>
>
> --
>
|
