FAQ
Hi,

We recently updated our web servers to Catalyst 5.9007 with Perl 5.12.

After the upgrade, we've consistently seen very high cpu on the machines > 90%. After much looking, it appears that the apache threads are stuck in Cache::Memcached disconnecting.

We use Catalyst:Plugin::Session::Store::Cache::Memcached and Plugin::Session. We've tried Cache::Memcached::Fast but see random segmentation faults.

So, I'm looking to see if anyone else has seen this problem or may have any suggestions?

One option we're considering is moving away from the Memcached session store and using DBIC instead? Any recommendations?

I appreciate any help.

Thanks,

Todd




-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.scsys.co.uk/pipermail/catalyst/attachments/20120216/7456f949/attachment.htm

Search Discussions

  • Toby Corkindale at Feb 17, 2012 at 1:27 am

    On 17 February 2012 11:16, Todd Benge wrote:
    Hi,

    We recently updated our web servers to Catalyst 5.9007 with Perl 5.12.

    After the upgrade, we've consistently seen very high cpu on the machines >
    90%. ?After much looking, it appears that the apache threads are stuck in
    Cache::Memcached disconnecting.
    Which Apache model are you using? Prefork, worker threads, events?
  • Todd Benge at Feb 17, 2012 at 1:48 am
    We're using PreFork.

    Sent from my iPhone
    On Feb 16, 2012, at 6:29 PM, Toby Corkindale wrote:
    On 17 February 2012 11:16, Todd Benge wrote:
    Hi,

    We recently updated our web servers to Catalyst 5.9007 with Perl 5.12.

    After the upgrade, we've consistently seen very high cpu on the machines >
    90%. After much looking, it appears that the apache threads are stuck in
    Cache::Memcached disconnecting.
    Which Apache model are you using? Prefork, worker threads, events?

    _______________________________________________
    List: Catalyst@lists.scsys.co.uk
    Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
    Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
    Dev site: http://dev.catalyst.perl.org/
  • Toby Corkindale at Feb 20, 2012 at 2:03 am
    Ah, I don't know then.

    Was just going to say that Cache::Memcached uses AnyEvent under the
    hood, and maybe that was interfering with apache's event loop if you
    were using that model.
    On 17 February 2012 12:48, Todd Benge wrote:
    We're using PreFork.

    Sent from my iPhone
    On Feb 16, 2012, at 6:29 PM, Toby Corkindale wrote:
    On 17 February 2012 11:16, Todd Benge wrote:
    Hi,

    We recently updated our web servers to Catalyst 5.9007 with Perl 5.12.

    After the upgrade, we've consistently seen very high cpu on the machines >
    90%. ?After much looking, it appears that the apache threads are stuck in
    Cache::Memcached disconnecting.
    Which Apache model are you using? Prefork, worker threads, events?

    _______________________________________________
    List: Catalyst@lists.scsys.co.uk
    Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
    Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
    Dev site: http://dev.catalyst.perl.org/
    _______________________________________________
    List: Catalyst@lists.scsys.co.uk
    Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
    Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
    Dev site: http://dev.catalyst.perl.org/


    --
    Turning and turning in the widening gyre
    The falcon cannot hear the falconer
    Things fall apart; the center cannot hold
    Mere anarchy is loosed upon the world
  • Vladimir Timofeev at Feb 20, 2012 at 5:08 am

    2012/2/17 Todd Benge <todd.benge@trackvia.com>:
    Hi,

    We recently updated our web servers to Catalyst 5.9007 with Perl 5.12.

    After the upgrade, we've consistently seen very high cpu on the machines >
    90%. ?After much looking, it appears that the apache threads are stuck in
    Cache::Memcached disconnecting.

    We use Catalyst:Plugin::Session::Store::Cache::Memcached and
    Plugin::Session. ?We've tried Cache::Memcached::Fast but see random
    segmentation faults.

    So, I'm looking to see if anyone else has seen this problem or may have any
    suggestions?
    I've seen similar problems when the connection (to memcached in this
    case) was set up before fork. After fork, you can't reuse same socket.
    For example: DBIC automatically handle it, by compare saved (on
    connect time) pid with current pid, and reconnect when they not equal.
    One option we're considering is moving away from the Memcached session store
    and using DBIC instead? ?Any recommendations?

    I appreciate any help.

    Thanks,

    Todd





    _______________________________________________
    List: Catalyst@lists.scsys.co.uk
    Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
    Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
    Dev site: http://dev.catalyst.perl.org/


    --
    Vladimir Timofeev <vovkasm@gmail.com>

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupcatalyst @
categoriescatalyst, perl
postedFeb 17, '12 at 12:16a
activeFeb 20, '12 at 5:08a
posts5
users3
websitecatalystframework.org
irc#catalyst

People

Translate

site design / logo © 2021 Grokbase