Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

storage system in consideration from below h/w lying idle: 6 x 300GB 10K SAS; 2 X 146 GB 15K SAS; 12GB DDR2 RAM; Xeon CPU 2.5Ghz two 1GB NIC ports 1 - DLink 1GB ethernet switch (1000/100 MB) 8-port unmanaged

Software: nexentastor community 3.0.3 or Opensolaris

target usage: iSCSI for ESX 4.0

what are the general recommendations for ZFS cache: read, write? 1. SATA/SAS or SSD?; 2. minimum recommended size?

any ideas to tune network? thanks in adv.

share|improve this question
    
If you want a video series where they talk about how ZFS works and some tuning goto youtube.com/user/deirdresun and it is called "ZFS in the Trenches: Ben Rockwood at LISA 09" (7 in all). Ben talked about turning the cache and how to monitor ZFS. His blog is at cuddletech.com/blog I'm not sure that bug is still active, it isn't posted as fixed, but a workaround is posted in a related bug. When you go to that bug it was fixed in snv_56. Not sure if the second bug fix, fixed the first bug. I don't remember a reference to memory leakage in the video series. –  Scott McClenning Oct 13 '10 at 3:10
add comment

1 Answer

You might not like this, but you might want to take a look at http://serverfault.com/questions/162693/my-opensolaris-server-hangs-when-writing-large-files-after-upgrading-zpool/163955#163955 if you haven't already...

share|improve this answer
    
do you mean this bug still exists in ZFS and hence not recommended for any VM storage solution (large file writes)? –  John-ZFS Aug 3 '10 at 2:10
    
It's not just large writes, it's a memory leak, large writes just seems to speed up the leakage... –  user48838 Aug 3 '10 at 2:43
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.