Hello, I have a failover Cluster with two XEN VMs on RHEL5 with a 3.6T partition from a SAN mapped as /dev/hdc in the VM. If I mount the device (/dev/hdc on /dot type gfs (rw,noatime,hostdata=jid=0:id=196610:first=1)) with lock_dlm at only one machine and go in production state, the filesystem performes well for about on day. After than it slows extremly down. If then I switch to the other clusternode or remount it performes well. "gfs_tool gettune /dot" says: ilimit1 = 100 ilimit1_tries = 3 ilimit1_min = 1 ilimit2 = 500 ilimit2_tries = 10 ilimit2_min = 3 demote_secs = 300 incore_log_blocks = 1024 jindex_refresh_secs = 60 depend_secs = 60 scand_secs = 5 recoverd_secs = 60 logd_secs = 1 quotad_secs = 5 inoded_secs = 15 quota_simul_sync = 64 quota_warn_period = 10 atime_quantum = 3600 quota_quantum = 60 quota_scale = 1.0000 (1, 1) quota_enforce = 1 quota_account = 1 new_files_jdata = 0 new_files_directio = 0 max_atomic_write = 4194304 max_readahead = 262144 lockdump_size = 131072 stall_secs = 600 complain_secs = 10 reclaim_limit = 5000 entries_per_readdir = 32 prefetch_secs = 10 statfs_slots = 64 max_mhc = 10000 greedy_default = 25 greedy_quantum = 6 greedy_max = 62 rgrp_try_threshold = 100
The mem ussage of "top" is: Mem: 4194304k total, 1202044k used, 2992260k free, 101972k buffers Swap: 1049592k total, 0k used, 1049592k free, 581156k cached Is there any tuning I coud do? Have tested "gfs_tool settune /dot demote_secs 500" and "gfs_tool settune /dot scand_secs 20" with no effects. I would be very pleased if anyone coud help me. Best regards, Dirk _______________________________________________ rhelv5-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/rhelv5-list
