search for: space_map_alloc

...anonpages+0x310 genunix`segvn_fault+0x438 genunix`as_fault+0x4c8 unix`pagefault+0x68 unix`trap+0xd4c unix`utl0+0x4c 71 genunix`avl_walk+0x20 zfs`metaslab_ff_alloc+0x9c zfs`space_map_alloc+0x10 zfs`metaslab_group_alloc+0x1e0 zfs`metaslab_alloc_dva+0x114 zfs`metaslab_alloc+0x2c zfs`zio_dva_allocate+0x4c zfs`zio_write_compress+0x1ec zfs`arc_write+0xe4 zfs`dbuf_sync+0x6c0 zfs`...

Hi. System is snv_56 sun4u sparc SUNW,Sun-Fire-V440, zil_disable=1 We see many operation on nfs clients to that server really slow (like 90 seconds for unlink()). It''s not a problem with network, there''s also plenty oc CPU available. Storage isn''t saturated either. First strange thing - normally on that server nfsd has about 1500-2500 number of threads. I did

...reds or thousands of small (~3KB) reads each second, presumably reading space map data from disk to find places to put the new blocks. The result is that it can take several minutes for an spa_sync() to complete, even if I''m only writing a single 128KB block. Using DTrace, I can see that space_map_alloc() frequently returns -1 for 128KB blocks. From my understanding of the ZFS code, that means that one or more metaslabs has no 128KB blocks available. Because of that, it seems to be spending a lot of time going through different space maps which aren''t able to all be cached in RAM at the...