Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hello Andreas, Thanks for responding, below is a section of my current sysctl.conf file. As you can see that I used the alternative parameters that you had sent me. The resource spikes vary from 18-24 hours apart. The odd thing is that it sometimes occurs when the system is just idling i.e. 5-18Mbytes a second. # # Kernel tweaks to close the gap between the buffercache and DRBD DIO to remote store. # vm.swappiness = 10 vm.zone_reclaim_mode = 1 vm.dirty_background_ratio = 1 vm.dirty_ratio = 1 vm.dirty_expire_centisecs = 300 vm.dirty_writeback_centisecs = 60 vm.vfs_cache_pressure = 40 The settings forces the delta between data coming into the file server and data being set to the peer to be 1-3 seconds as opposed to nearly 25seconds. -----Original Message----- From: drbd-user-bounces at lists.linbit.com [mailto:drbd-user-bounces at lists.linbit.com] On Behalf Of Andreas Kurz Sent: Tuesday, December 18, 2012 4:30 AM To: drbd-user at lists.linbit.com Subject: Re: [DRBD-user] Still experiencing resource spikes On 12/17/2012 08:27 PM, Prater, James K. wrote: > > > > > > > I have a real sticky problem and would appreciate if anyone has insight. > > > > > > We currently have the following physical configuration > > > > 2) Dell PowerEdge R710 (dual 6-core with hyperthreading enabled) > > 120Gbytes of memory each That's quite a lot ... you also tuned vm.dirty_background_bytes and vm.dirty_bytes to a reasonable low value? .... to avoid regular heavy data write-out.... Regards, Andreas -- Need help with DRBD? http://www.hastexo.com/now > > 4x 10GE Nics (2-bonded for NFS and 2-bonded for Replication) > Broadcom 57711. > > 4x 8Gbit FC HBAs which are tied to a dual controller NEXSAN E60 > array (controllers in non-redundant mode), each controller has 4Gbytes > of memory. > > Raw throughput is around 340Mbytes/sec per volume. > > > > Each system is running RHEL 6.3 with heartbeat and now with > DRBD-8.4.2, problem was there with 8.4.0 (could not get 8.4.1 to work). > > System configured as Active/Passive pair with EXT4 and the filesystem, > barriers off. Filesystems exported via NFS to vSphere 4.1 clients. > > > > Main problem is that everything works for most of the time but every > now and then a resource stall (high load average and no I/O) occurs > which > > Is not good for running VMs. Has anyone seen this? No errors > recorded just no I/O and high load for a few minutes (3-4). This has > been driving > > me crazy. One more thing, these events do not occur with > "replication disabled" i.e. drbdadm down all (on the peer member). > I have adjusted > > many sysctl parameters (up memory buffers , etc) , changed I/O > schedulers and turned on and off hyperthreading and still have the issue. > > > > > > Thanks in advance. > > > > > > James > > > > > > > > > > > > > > > > _______________________________________________ > drbd-user mailing list > drbd-user at lists.linbit.com > http://lists.linbit.com/mailman/listinfo/drbd-user >