Software-based VM-centric and flash-friendly VM storage + free version
Moderators: anton (staff), art (staff), Max (staff), Anatoly (staff)
-
MichelZ
- Posts: 34
- Joined: Sun Mar 16, 2014 10:38 am
-
camealy
- Posts: 77
- Joined: Fri Sep 10, 2010 5:54 am
Fri Apr 03, 2015 2:37 pm
Far from it. I can only get decent performance from LSFS if L1 and L2 cache are disabled. I submitted an email about one of my HA nodes completely dropping two devices with just a reboot. And now every time I restart either node I lose portions of my initiator access rules.
-
Oles (staff)
- Staff
- Posts: 91
- Joined: Fri Mar 20, 2015 10:58 am
Mon Apr 06, 2015 6:44 pm
Thank you for the information, we are currently in the middle of investigation. We will provide an update as soon as we hear back from the customer.
-
camealy
- Posts: 77
- Joined: Fri Sep 10, 2010 5:54 am
Wed Apr 08, 2015 1:21 pm
I have sent in all logs requested multiple times. So I don't believe you are waiting on me for anything.
I would offer a word of caution to anyone considering the latest build of Virtual SAN. We currently lose multiple access rules upon service restart or node reboot, causing iSCSI initiators in our cluster members to mark the targets offline (due to rejection instead of being unavailable and in a reconnecting state). We also lose devices from nodes in HA sets randomly during the same service restart or server reboot. (not limited to LSFS, thick images as well) These problems are in addition to LSFS L1 cache performance issues.
-
Oles (staff)
- Staff
- Posts: 91
- Joined: Fri Mar 20, 2015 10:58 am
Fri Apr 10, 2015 11:29 am
Community update,
We are currently investigating your logs, I will inform you as soon as we get any updates.
Thank you for your patience.
-
MichelZ
- Posts: 34
- Joined: Sun Mar 16, 2014 10:38 am
-
camealy
- Posts: 77
- Joined: Fri Sep 10, 2010 5:54 am
Fri Apr 24, 2015 11:38 am
They have kept me updated with status, but fixes are still in development. I have taken the system out of production while we wait.
I will be extremely nervous to trust the system with production loads going forward though.
-
MichelZ
- Posts: 34
- Joined: Sun Mar 16, 2014 10:38 am
Fri Apr 24, 2015 11:51 am
Yeah, me too. It's a pitty... this has been good software in the past. Seems like it's falling apart
-
camealy
- Posts: 77
- Joined: Fri Sep 10, 2010 5:54 am
Fri Apr 24, 2015 12:00 pm
I agree, most builds of v6 were reliable as an anvil. This v8 reminds me of 6 years ago when they announced 5.0 HA. We became Starwind partners and made the mistake of adopting that right away and lost an entire production Exchange system.
-
Anatoly (staff)
- Staff
- Posts: 1675
- Joined: Tue Mar 01, 2011 8:28 am
-
Contact:
Fri Apr 24, 2015 5:53 pm
Hi guys.
I know that you heared that million times alreaday, but please take my apologies for your situations.
We will do our best to keep you happy. Take my word for that.
Best regards,
Anatoly Vilchinsky
Global Engineering and Support Manager
www.starwind.com
av@starwind.com
-
MichelZ
- Posts: 34
- Joined: Sun Mar 16, 2014 10:38 am
Mon Apr 27, 2015 9:58 am
Are these L1/L2 cache performance issues only related to LSFS, or to normal HA-images as well?
-
Anatoly (staff)
- Staff
- Posts: 1675
- Joined: Tue Mar 01, 2011 8:28 am
-
Contact:
Tue Apr 28, 2015 10:44 am
OK, lets clarify:
The L1 caching works great right now.
The L2 cache will work great with next update.
Best regards,
Anatoly Vilchinsky
Global Engineering and Support Manager
www.starwind.com
av@starwind.com