Hi, we tried building an storage server ourself. For that we took a Tyan S7012 mainboard, 72 GB of memory and one Adaptec 52445 and one Adaptec 51645 to attach 38 disks. Two system disks are directly attached to the motherboard.
After we replaced the SUNWaac driver with the one from adaptec we were able to see all disks and we created one big zpool for testing purposes. During the tests we did some read/write-intensive workloads (copy one 30 GB file three times in parallel, extract the linux kernel, remove everything and start over again). After some time I/O to the disks attached via both adaptec controller (38 disks total) stalls. No I/O can be made. Even an "ls /pool1" stalls. However I/O to the rpool (system disks are directly attached to the motherboard) still works. In order to find out where the problem might be located i created two full page core dumps on two different I/O hang occasions. For analyzing them i need some help. -- Ronny Egner -- This message posted from opensolaris.org _______________________________________________ storage-discuss mailing list storage-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/storage-discuss
