Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
>>>>When I only leave one node(primary) I still have laga. And it is >>>>strangest thing to me. It is x86_64 HA Cluster, but I have the same >>>>hardware in another location and there are no problems. Strange :) >>> >>>right. so I'll file that as "somebody elses problem", >>>as it seems not to be drbd problem. >> >>We have changed almost everything. >>1. have tested kernels from 2.6.14 to 2.6.17 >>2. drbd from 0.7.14 to 0.7.21 >>3. changed machine to different vendor >>4. changed SCSI disks to IDE >>5. changed x86_46 architecture to i386 >>And all fail. All the time we get console lags. >>The strangest thing is that when we run mysql on phisical >>device(/dev/sdaX) there is no lags. >>I don't know how can I debug it. > > > I'm certainly not an expert in reading those kernel traces, but even for > non-kernel hackers it may give an idea whats going on. Also, the nice linbit > people certainly know how to read it. > > One can get a trace of the current tasks via 'sysrq t' , > see /usr/src/linux/Documentation/sysrq.txt for details. Hello, Thank You verry much for Your help. The problem was trivial and it wasn't drbd. It was heartbeat, exactly we set SERVICE_RUN_NICE_LEVEL="-12" in /etc/sysconfig/heartbeat and mysql consumed all resoures, and we got lags. Best Regards Maciej Bogucki