Hi,
Incremental GC has nothing to do with it.  "Rebuilding the cache"
(whatever that means) has nothing to do with it.  This has been
discussed numerous times in the past, and has to do with the permanent
generation as you've surmised.  See for example:
http://marc.theaimsgroup.com/?l=tomcat-user&m=107590242329345&w=2
http://marc.theaimsgroup.com/?l=tomcat-user&m=107996450114197&w=2
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=20758
http://marc.theaimsgroup.com/?l=tomcat-user&m=107048670308534&w=2

Ahh, I'm tired of searching.

Yoav Shapira
Millennium Research Informatics


>-----Original Message-----
>From: Angelov, Rossen [mailto:[EMAIL PROTECTED]
>Sent: Wednesday, April 07, 2004 9:39 PM
>To: 'Tomcat Users List'
>Subject: RE: Tomcat 5 out of memory
>
>That sounds like a reasonable explanation.
>
>Then, the only solution I can see is to set the development environment
not
>a virtual server per developer but a Tomcat instance per person and if
the
>OutOfMemory error happens again, restarting that Tomcat instance won't
>interfere with the rest of the developers' work.
>
>Do you by any chance have a link to that discussion you mentioned?
>I searched the archive and found some opinions, the closest was this
one
>"...common classloader repositories (as well as shared and server) do
not
>get discarded on webapp reload.  They can't, as that might destroy
other
>webapps."
>
>This probably raises the question in what cases the reload should be
used.
>
>Thanks,
>Ross
>
>-----Original Message-----
>From: Parsons Technical Services
[mailto:[EMAIL PROTECTED]
>Sent: Wednesday, April 07, 2004 7:54 PM
>To: Tomcat Users List
>Subject: Re: Tomcat 5 out of memory
>
>
>Ross,
>
>If my memory serves me right (which is rarely) there was a discussion
on
>the
>list about this. I believe the comment was that during reloads there
are
>references to objects that don't get released when a context is
undeployed.
>The memory is lost until a restart. In production this should not be an
>issue because of the infrequent occurrence of redeploys.
>
>Doug
>www.parsonstechnical.com
>
>----- Original Message -----
>From: "Angelov, Rossen" <[EMAIL PROTECTED]>
>To: "'Tomcat Users List'" <[EMAIL PROTECTED]>
>Sent: Wednesday, April 07, 2004 7:13 PM
>Subject: RE: Tomcat 5 out of memory
>
>
>> I don't understand why rebuilding the cache would take more memory
than
>what
>> was originally required. It just doesn't sound right or it is a
memory
>leak
>> when you use the stop/start or reload options.
>>
>> Giving a higher priority to this process is not a very good option
>because
>> in our case it is a development UNIX server and there are database
and
>other
>> peoples' processes running.
>>
>> Increasing the memory heap size will just slow down the occurrence of
>that
>> OutOfMemory error instead of preventing it. I'm actually looking for
a
>long
>> term solution because it may be critical whether to use tomcat for
>> production and live web servers.
>>
>> Thanks,
>> Ross
>>
>> -----Original Message-----
>> From: Nicholas Bernstein [mailto:[EMAIL PROTECTED]
>> Sent: Wednesday, April 07, 2004 5:29 PM
>> To: Tomcat Users List
>> Subject: Re: Tomcat 5 out of memory
>>
>>
>> rebuilding the cache, I would imagine. I'm not sure how you would
reduce
>> it though, it's "a good thing"(tm) -- really, it's reducing the
memory
>> usage by 2-3MB once the cache is rebuilt.
>>
>> This is just an (somewhat) educated guess, though. I suppose you
could
>> give it a higher priority so it could speed it up, and take up more
>> resources, for a shorter period of time, or alternately, throw some
>> hardware at it, and add some more memory.
>>
>> On Wed, 2004-04-07 at 15:17, Angelov, Rossen wrote:
>> > Hi,
>> >
>> > We are having problems with Tomcat when restarting the virtual
servers
>or
>> > the contexts - the memory usage goes up with about 2-3 MB per
restart.
>> > We have the following java options when starting Catalina:
>> > -Xms128m -Xmx256m -verbose:gc
>> >
>> > Does anybody know what's causing such a behavior and how it can be
>> limited?
>> >
>> > Thanks,
>> > Ross
>> >
>> > "This communication is intended solely for the addressee and is
>> > confidential and not for third party unauthorized distribution."
>> --
>> +---------------------------------------------------------------+
>> | Nicholas Bernstein            | [EMAIL PROTECTED]             |
>> | UNIX Systems Administrator    | http://www.docmagic.com       |
>> | Document Systems Inc.         |  |
>> | gpg: F706 8C4E 78FA DDDD 53A0 019F D983 FE28 2002 D1F3 |
>> +---------------------------------------------------------------+
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [EMAIL PROTECTED]
>> For additional commands, e-mail: [EMAIL PROTECTED]
>>
>> "This communication is intended solely for the addressee and is
>> confidential and not for third party unauthorized distribution."
>>
>>
>
>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: [EMAIL PROTECTED]
>For additional commands, e-mail: [EMAIL PROTECTED]
>
>
>"This communication is intended solely for the addressee and is
>confidential and not for third party unauthorized distribution."




This e-mail, including any attachments, is a confidential business communication, and 
may contain information that is confidential, proprietary and/or privileged.  This 
e-mail is intended only for the individual(s) to whom it is addressed, and may not be 
saved, copied, printed, disclosed or used by anyone else.  If you are not the(an) 
intended recipient, please immediately delete this e-mail from your computer system 
and notify the sender.  Thank you.


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to