Hello
Have any of you experienced that P2V takes a long time? We are running ESXi 4.0, 244038 on Dell R710.
A P2V of a Windows 2003 server whit one 70 GB disk takes 12 hours.
I have not changed anything regarding disks, RAM, CPU or disk size. It is a 1 to 1.
My ESXi host are taged til all my VLAN on my HP network.
I get som entries in my log:
Apr 12 09:23:50 10.28.4.167 Hostd: [2011-04-12 09:23:50.017 46249DC0 warning 'Proxysvc'] SSL Handshake failed for stream TCP(local=AAA.AAA.AAA.AAA:55419, peer=BBB.BBB.BBB.BBB:443), error=SSL Exception: error:14094412:SSL routines:SSL3_READ_BYTES:sslv3 alert bad certi
AAA.AAA.AAA.AAA = is the IP of the server where i an running the stand alone convoter version 4.3 from, and BBB.BBB.BBB.BBB is the IP of the ESXi 4.0 host.
Are both the ESX host and the physcial machine on the same network segment? What are the respective network speeds for each machine?
Hi,
If you are on a decent network connection, then 70GB should not take 12 hours (though it can)
More often than not, I have found that long P2V times are due to one of 3 things:
Hello weinstein5
The host ant the physcial machine are not on the same network segment and have never been. The physcial machine have a 100 Mb NIC and i set to auto. The host have 1000 Mb network.
As others have posted - I think the slowness is due to the 100 MB network - are the two machines in the same datacenter? If not and the traffic is traversing the WAN then that too will slow transfer speeds -
HI bulletprooffo…
Yes the to serveres are in the same datacenter and serverrum using the same switch. It is properly the 100 Mb NIC......
I'd try disabling the AV for starters.
Also, at 100MB . . the absilute best time you'd get for a flat file copy is almost 2 hours - considering this is a P2V . . I am not surprised that it is taking longer.
I would imagine you'd get better tiomes than 12 hours . . but at the same time, I am not surprised.
I'd imagine with some tweaking / dedicated storage paths, single copy running at a time etc, you'll be able to shave a few hours . . but I wouldn't expect to see much less than about 4 hours per P2V.
Just one other note. I have found if physical servers have like millions of small files on them, that the P2V slows down a lot.
We tested a difficult P2V last weekend and it looked like it was about 16 hours per TB, just for comparisons. 1GIG lan, 1gig cards in host and phycial SVR, same segment / same switch / same subnet.
Hello
After talking with VMware support they found out that inconsistent pathing causes path thrashing which in turn reduces storage performance. This was only in this one volume. All the other volume is ok.
Hi cawn - have you now resolved this issue then?
How long did the P2V take once the amendment was made, and could you post more details on the fix, then mark the thread as answered for future readers with matching issues please?
Thanks
Hello.
It requires that this volume is set to RR and not fixed. And a complet Rescan og the HBA's. I have not done this yet, I'm missing a service window. But when this i don, I vill poste back to you all whit the result.
Path thrashing... Thanks for the update. Something new to me that I probably should know about.
It has helped to put the path to RR and do a Rescan.
Could not take the same server, but took a similar server.
It was done in 8 hours and not 20 hours.
But still long time.