On Mon, 20 Jun 2016 01:10:22 +0200 Jonathan Aquilina <jaquil...@eagleeyet.net>
said:

> Hi Carsten, 
> 
> What virtualization platform is all this running on? 

qemu+kvm+libvirt

> On 2016-06-13 09:27, Carsten Haitzler wrote:
> 
> > On Fri, 10 Jun 2016 20:24:19 +0200 Jonathan Aquilina
> > <jaquil...@eagleeyet.net> said:
> > 
> >> If there is build servers used heavily for builds I am more then willing
> >> to host the site no problem.  Beber gave me a project I hope to try and
> >> figure out with the planet this weekend other then that I cant
> >> investigate further this issue of the server that is crashing. 
> >> 
> >> @Beber not sure if you can give me access to the server in question or
> >> even Raster as I have a hunch that somehow either a hardware fault or
> >> out of memory issue that the web server is running into I would need to
> >> see the logs.
> > 
> > i don't know if this was the issue this time but the general issue where i
> > end up rebooting the vm host every now and again is this:
> > 
> > systems run for some time. the GUEST vm's at some point grind to a halt. not
> > quite a halt but it seems like any i/o causes some timeout. like "ls" on the
> > cmdline in a guest can take between 30 to 150 seconds. it doesn't spin using
> > cpu. it just hangs. it's like a timeout of some sore, then finally it works.
> > the system is not overloaded. the host machine is pretty much idle. as are
> > the vm's. you run top in one shell-in and do the ls in another - nothing is
> > loaded either there. it's as if i/o is doing something like logging and a
> > dns lookup or a connect is timing out.
> > 
> > now i have tried simply shutting down the vm in question and starting it
> > again
> > - it stays "bad". i've shut down all vm's and started them again (it takes
> > forever thanks to the i/.o timeout issue) and nothing is fixed. the only
> > thing that fixes it is a reboot of the host os that brings all vm's down,
> > reboots the host os then starts them up and all is fine again... until the
> > next time this happens. i have rummaged around logs on the host and guests.
> > i have looked at every dmesg and more and i couldn't find anything telling
> > me what was going on. maybe i missed something but i only have so much time
> > to spend on these things before the simple "reboot the sucker" option needs
> > to be used and we have work to get on with.
> > 
> > it's not because of a build server overload - nothing is overloaded. fyi the
> > HOST os works just fine. no langs. i/o works perfectly. i just don't know
> > what it is. :(
> > 
> > On 2016-06-10 18:45, Cedric BAIL wrote:
> > 
> > On Fri, Jun 10, 2016 at 8:26 AM, Jonathan Aquilina
> > <jaquil...@eagleeyet.net> wrote: 
> > 
> > Where is this being hosted currently? If you guys want I can host a
> > backup site on my server. Only thing I would not be able to host are the
> > large amount of download tarballs etc. 
> > I don't think it is an hosting issue at this point. We do already have
> > 3 operationals servers as I know. One master in the US (the biggest
> > one doing the build), two in france (One able to run test on GPU and
> > the other being a dev server). The current problem is I think one that
> > has been lurking around for some time. Once in a while the server goes
> > overloaded and we haven't had the opportunity to investigate what is
> > going on. Raster usually just reboot the server and that solve this. I
> > have asked Beber to look at it so that we can fix the problem instead
> > of requiring a reboot every n months.
> > 
> > Cedric
> > 
> > On 2016-06-10 16:00, Amitesh Singh wrote:
> > 
> > Hello Guys,
> > 
> > It seems like all web stuffs of e.org are down. Also git clone/pull/push is
> > super slow.
> > 
> > please check.
> > 
> > Regards 
> > ------------------------------------------------------------------------------
> > What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
> > patterns at an interface-level. Reveals which users, apps, and protocols are
> > consuming the most bandwidth. Provides multi-vendor support for NetFlow,
> > J-Flow, sFlow and other flows. Make informed decisions using capacity
> > planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
> > _______________________________________________
> > enlightenment-devel mailing list
> > enlightenment-devel@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
> 
> ------------------------------------------------------------------------------
> What NetFlow Analyzer can do for you? Monitors network bandwidth and
> traffic
> patterns at an interface-level. Reveals which users, apps, and protocols
> are 
> consuming the most bandwidth. Provides multi-vendor support for NetFlow,
> 
> J-Flow, sFlow and other flows. Make informed decisions using capacity 
> planning reports.
> https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
> _______________________________________________
> enlightenment-devel mailing list
> enlightenment-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
>  
> ------------------------------------------------------------------------------
> What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
> patterns at an interface-level. Reveals which users, apps, and protocols are 
> consuming the most bandwidth. Provides multi-vendor support for NetFlow, 
> J-Flow, sFlow and other flows. Make informed decisions using capacity planning
> reports. http://sdm.link/zohomanageengine
> _______________________________________________
> enlightenment-devel mailing list
> enlightenment-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
> 


-- 
------------- Codito, ergo sum - "I code, therefore I am" --------------
The Rasterman (Carsten Haitzler)    ras...@rasterman.com


------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are 
consuming the most bandwidth. Provides multi-vendor support for NetFlow, 
J-Flow, sFlow and other flows. Make informed decisions using capacity planning
reports. http://sdm.link/zohomanageengine
_______________________________________________
enlightenment-devel mailing list
enlightenment-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel

Reply via email to