Re: [ceph-users] threading requirements for librbd

2016-03-10 Thread Blair Bethwaite
Thanks Jason, thanks Dan, On 9 March 2016 at 01:34, Jason Dillaman wrote: > Are you interesting in the max FD count or max thread count? You mention > both in your email. True, I did mix the two somewhat incorrectly - I was sort of guessing there'd be some number of threads per socket or vice

Re: [ceph-users] threading requirements for librbd

2016-03-08 Thread Jason Dillaman
hwaite" > To: ceph-users@lists.ceph.com > Sent: Tuesday, March 8, 2016 7:32:01 AM > Subject: [ceph-users] threading requirements for librbd > > Hi all, > > Not getting very far with this query internally (RH), so hoping > someone familiar with the code can spare me

Re: [ceph-users] threading requirements for librbd

2016-03-08 Thread Dan van der Ster
Hi Blair! Last I heard you should budget for 2-3 fds per OSD. This only affects Glance in our cloud -- the hypervisors run unlimited as root. Here's our config in /etc/security/limits.d/91-nproc.conf: glance softnofile 32768 glance hardnofile 32768 glance softnpro

[ceph-users] threading requirements for librbd

2016-03-08 Thread Blair Bethwaite
Hi all, Not getting very far with this query internally (RH), so hoping someone familiar with the code can spare me the C++ pain... We've hit soft thread count ulimits a couple of times with different Ceph clusters. The clients (Qemu/KVM guests on both Ubuntu and RHEL hosts) have hit the limit th