On Jan 29, 2013, at 11:01 AM, seaside-request@lists.squeakfoundation.org wrote:
From: Esteban Lorenzano estebanlm@gmail.com Subject: Re: [Seaside] my site is completely dead.. Date: January 29, 2013 11:01:04 AM GMT+01:00 To: Seaside - general discussion seaside@lists.squeakfoundation.org Reply-To: Seaside - general discussion seaside@lists.squeakfoundation.org
it depends on many things... as Sebastian said, nothing in the earth runs forever, but you can achieve some stability anyway :)
first I would check your Pharo version and VM version, to ensure you are using a stable one. second, RFB server (vnc server) is known for hang your image time to time (some sockets stays alive when they shouldn't and that is consuming your resources until nothing works :( ). Someone should fix that, but is obviously not so easy, otherwise it would be fixed by now.
what else? ensure your image is being persisted and backed up regularly, and add a monitoring process...
Esteban
The issue in RFB was identified and a fix posted, don't think any RFB package maintainers picked it up though. http://forum.world.st/leaking-semaphores-tp4432983p4433697.html
As for Pharo, the relevant issue to giving better error messages instead of freezing is http://code.google.com/p/pharo/issues/detail?id=5423 but I never got around to actually implementing it :S
Also, a new(ish) VM will happily continue leaking semaphores instead of freezing till it runs out of space just the way non-cog used to, if I've understood Igor's latest changes correctly :)
Cheers, Henry
ok, I rescued the issue from oblivion, but it will take some time until I can work on it. If someone wants to produce a slice or a changeset (better a slice), I will be happy to review and integrate :)
Esteban
On Jan 29, 2013, at 3:26 PM, Henrik Johansen henrik.s.johansen@veloxit.no wrote:
On Jan 29, 2013, at 11:01 AM, seaside-request@lists.squeakfoundation.org wrote:
From: Esteban Lorenzano estebanlm@gmail.com Subject: Re: [Seaside] my site is completely dead.. Date: January 29, 2013 11:01:04 AM GMT+01:00 To: Seaside - general discussion seaside@lists.squeakfoundation.org Reply-To: Seaside - general discussion seaside@lists.squeakfoundation.org
it depends on many things... as Sebastian said, nothing in the earth runs forever, but you can achieve some stability anyway :)
first I would check your Pharo version and VM version, to ensure you are using a stable one. second, RFB server (vnc server) is known for hang your image time to time (some sockets stays alive when they shouldn't and that is consuming your resources until nothing works :( ). Someone should fix that, but is obviously not so easy, otherwise it would be fixed by now.
what else? ensure your image is being persisted and backed up regularly, and add a monitoring process...
Esteban
The issue in RFB was identified and a fix posted, don't think any RFB package maintainers picked it up though. http://forum.world.st/leaking-semaphores-tp4432983p4433697.html
As for Pharo, the relevant issue to giving better error messages instead of freezing is http://code.google.com/p/pharo/issues/detail?id=5423 but I never got around to actually implementing it :S
Also, a new(ish) VM will happily continue leaking semaphores instead of freezing till it runs out of space just the way non-cog used to, if I've understood Igor's latest changes correctly :)
Cheers, Henry _______________________________________________ seaside mailing list seaside@lists.squeakfoundation.org http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside
seaside@lists.squeakfoundation.org
-
Esteban Lorenzano -
Henrik Johansen