Re: OOM killer invoked during btrfs send/recieve on otherwise idle machine

2016-08-19 Thread Michal Hocko
On Tue 02-08-16 09:48:01, Chris Murphy wrote: > On Tue, Aug 2, 2016 at 9:38 AM, Chris Murphy wrote: > > Yesterday I saw oom killer knocking off processes during a simple cp > > -a from one Btrfs to another, in a VM, with kernel > > 4.8.0-0.rc0.git3.1.fc25.x86_64. The call trace looks different tha

Re: OOM killer invoked during btrfs send/recieve on otherwise idle machine

2016-08-02 Thread Chris Murphy
On Tue, Aug 2, 2016 at 9:38 AM, Chris Murphy wrote: > Yesterday I saw oom killer knocking off processes during a simple cp > -a from one Btrfs to another, in a VM, with kernel > 4.8.0-0.rc0.git3.1.fc25.x86_64. The call trace looks different than > Markus' so I'm not sure it's the same problem. It'

Re: OOM killer invoked during btrfs send/recieve on otherwise idle machine

2016-08-02 Thread Chris Murphy
Yesterday I saw oom killer knocking off processes during a simple cp -a from one Btrfs to another, in a VM, with kernel 4.8.0-0.rc0.git3.1.fc25.x86_64. The call trace looks different than Markus' so I'm not sure it's the same problem. It's not always reproducible. It hasn't happened on 4.7.0 though

Re: OOM killer invoked during btrfs send/recieve on otherwise idle machine

2016-08-02 Thread Michal Hocko
JFYI http://lkml.kernel.org/r/20160801192620.gd31...@dhcp22.suse.cz sounds quite similar to your report. order-2 OOMs with btrfs and many pagecache pages wrt. to the anon. I suspect that btrfs is preventing the compaction for some reason. On Sun 31-07-16 17:25:22, Markus Trippelsdorf wrote: > On 2

Re: OOM killer invoked during btrfs send/recieve on otherwise idle machine

2016-07-31 Thread Markus Trippelsdorf
On 2016.07.31 at 17:10 +0200, Michal Hocko wrote: > [CC Mel and linux-mm] > > On Sun 31-07-16 07:11:21, Markus Trippelsdorf wrote: > > Tonight the OOM killer got invoked during backup of /: > > > > [Jul31 01:56] kthreadd invoked oom-killer: > > gfp_mask=0x27000c0(GFP_KERNEL_ACCOUNT|__GFP_NOTRACK

Re: OOM killer invoked during btrfs send/recieve on otherwise idle machine

2016-07-31 Thread Michal Hocko
[CC Mel and linux-mm] On Sun 31-07-16 07:11:21, Markus Trippelsdorf wrote: > Tonight the OOM killer got invoked during backup of /: > > [Jul31 01:56] kthreadd invoked oom-killer: > gfp_mask=0x27000c0(GFP_KERNEL_ACCOUNT|__GFP_NOTRACK), order=2, oom_score_adj=0 This a kernel stack allocation. >

OOM killer invoked during btrfs send/recieve on otherwise idle machine

2016-07-30 Thread Markus Trippelsdorf
Tonight the OOM killer got invoked during backup of /: [Jul31 01:56] kthreadd invoked oom-killer: gfp_mask=0x27000c0(GFP_KERNEL_ACCOUNT|__GFP_NOTRACK), order=2, oom_score_adj=0 [ +0.04] CPU: 3 PID: 2 Comm: kthreadd Not tainted 4.7.0-06816-g797cee982eef-dirty #37 [ +0.00] Hardware name: