V6 Direct SAN Mode: Unable to allocate processing resources

[WayneH]

I've filled out a support request but did not include the SAN type, etc.

For those interested, we are having the issue with EMC VNX5300's connected via 10Gb iSCSI (Intel X520-DA2 NICs and Cisco 5548 switches). Veeam is running on a Dell R510 with 2008 R2 with PowerPath installed. Sometimes SAN mode works but the majority of time it uses the network.

[pinkerton]

btw it is just great that your're keeping us up to speed here. Most companies structures are so complex today and it is so diffcult to get information. I really like how Veeam handles this.

[Gostev]

No problems. Sorry for delay - the problem we had today is that none of pilot customers for whom the pilot patch did NOT work (for most, it helped as expected) were able to work with us to troubleshoot, so we lost much time for nothing. I even suggested to support that I could ask on forum for help... but they have at least one customer available now.

Guess this is what happens when the issue is not so critical

[Gostev]

If anyone having this issue is available today and/or tomorrow to do a webex with our support to validate the fix, please open a support case and include the link to this topic a part of your submission text as well as preferred time. Thanks!

Code: Select all

http://forums.veeam.com/viewtopic.php?f=24&t=9535 

[EdwardPoll]

No one has tried to contact us today....via phone, email or forums.....wasted day all round then, as I've just sat here waiting for Veeam to contact me....grrrr

Gostev, can you get support or someone to contact us, we are more than willing to help, tomorrow if not today..

Our outstanding ticket id is ID#5158692

Thanks

[Gostev]

Edward, it sounds like you may have wrong expectations. Most affected customers should not expect to be contacted by our support until the fix is validated. This is normal support process, we cannot be distributing unvalidated fix widely.

Thanks for your offer to help validating the fix, I have forwarded your support case ID to QC and they are calling you right now - please pick up the phone

[Gostev]

OK, so we need more pilot customers now, since the fix does help in the Edward's environment. We are still trying to reach those two for whom the fix did NOT help - this would be ideal - hope they will read this.

[aweiss]

Anton, let me know if you need, i leave my customer site but i kept a VPN access to be able to apply the fix ASAP. By the way, I also opened a new case for problem regarding VSS Freeze Time out since upgrade. (SQL won't freeze after v6 upgrade) Rgds Alex

[ph0en1x1]

[merged]

We have multiple EMC Sans and backup around 250 VM's per night totalling 20TB. We had the issue whereby the backups were not being performed direct from the San but would always fail over to the Lan. This made our backups impossible because of the data concerned and the Lan speed. We followed the supposed solution to this in the sticky by manually selecting our datastores but this made no difference. We were then informed that there would be an imminent fix for this.

After waiting we are now told that the fix is not immenent and there is no ETA.

We have had to uninstall V6 and revert to V5

My question for the developers is " Why was this basic functionality tested before release before allowing your customers to upgrade in a production environment"

Regards

[Gostev]

Alexandre, what is the case ID for SAN issue?

We are at the limited release stage for this patch right now, testing it on select customers. So far, have fixed the issue for all 4 who applied, including one of those who previously reported that the same patch did not help (probably due to environmental issue). If everything keep going well with a few more pilot deployments, we will be able to do the public release tomorrow.

[Gostev]

Mark, what is your support case ID? We are interested to try out the patch on various SAN models, since it appeared to be storage specific issue. As far as I know, some EMC SAN are indeed affected.

My question for the developers is " Why was this basic functionality tested before release before allowing your customers to upgrade in a production environment"

Because we do not have every possible SAN make and model in our labs. We did have 2 months long private beta with many existing customers, however apparently even that was not enough to cover all possible models.

Thanks!

[Jamie]

Hi,
I have this issue and am happy to try a 'beta' patch
SAN is a Clariion CX-320
Case 5159001

Cheers,

Jamie

[EdwardPoll]

Hi guys....yep can confirm it works for us, last night we stopped the Veeam proxies from failing back to NBD and we are 100% successful this morning.

For comparison sake, our environments is........

2 x HP EVA8400 running XCS 10000000
2 x EMC NS480 (CX4-480) connected over FC running Flare 6.0.41.4 on the Control Station with Clariion Code 04.30.000.5.517
HP Blades (BL460C)
Brocade fabric
vSphere 5.0.0,504890

Oh and the vStorage API is manually disabled in vSphere when talking to the Clariions (CX4-480) because of a bug/imcompatibility between vsphere 5 and the curently releases of Flare.

Thanks

Ed

[aweiss]

Alexandre, what is the case ID for SAN issue?

We are at the limited release stage for this patch right now, testing it on select customers. So far, have fixed the issue for all 4 who applied, including one of those who previously reported that the same patch did not help (probably due to environmental issue). If everything keep going well with a few more pilot deployments, we will be able to do the public release tomorrow.

Anton : case is 5159136 (SAN Direct Access)
Case for VSS issu is 5159575
Alex

[aweiss]

About which SAN, i have CX4-120 and a collegue runs with Compellent SC40 (Dell)

[cake]

Hello,

When i started an Backup Job it uses the NBD mode.

"09.12.2011 11:25:01 :: Using source proxy VMware Backup Proxy [nbd]"

Why?

Where do i get this Patch, or is this normal?
We have 3 x EMC CX4-480 and 1 x EMC CX4-240
Backup Server is a Fujitsu RX300 with 2 HBA.
The Backup LUN is 30TB.
The Backup see all LUNs from the ESXi Server
vsphere 5.0.0
Brocade Fibre Switch for the SAN.
HP Switch for Production and Backup LAN


Thx Carsten

[Gostev]

Hi Carsten, please read this topic, as well as sticky "before you upgrade" topic.

[aweiss]

Hi Guys
Veeam Support just phoned me to connect via webex. Hotfix1 is applyed and san direct acces problem is now resolved on a test job. I'll wait to see with other job tonight but should be ok. Still waiting for a fix on VSS Timeout error with SQL
Thks Anton and also to your team even if we would all wishes not having these troubles

[fnbuser]

We are having this problem as well. Support ticket ID#5159797.

Our SAN is a Clariion CX320.

I posted a new topic about this problem a little while ago (which may be stuck in moderation) before seeing this thread, so I apologize for that.

-Jason

[Gostev]

Hi Jason, our support started sending the official patch to all affected customers just a few hours earlier - so you should expect to be contacted soon. Thanks!

[WinstonWolf]

[merged]

Hello ,
Some words about my environment ,
My Backup & Replication 6 proxy is an W7 64bit Computer ( IntelCore I5 , 4 GB RAM ) , he Backups to an ISCSI Target ( Qnap TS 419 NAS ) over ISCSI Initiator from Windows.
My virtual Machines vmdk Files are stored on another Qnap ( TS 459 NAS ) which is connected via ISCSI to an ESX 4.1 U2 Server .
On my ESX Server i have an Vcenter Server Appliance as an VM running.

Now my first question , should i say that my vcenter is my Server or should i choose the ESX Server himself ?

And now to my Problem ,
When i backup my virtual Machines at the Moment over the ESX Server himself i become the follow Warning message :

"Unable to establish direct connection to the shared storage (SAN). Please ensure that: - HBA is properly installed in the Veeam Backup server computer, or software iSCSI initiator is configured correctly. - SAN volume can be seen by operating system in the Windows Disk Management snap-in on the Veeam Backup server. - Read access is allowed for the Veeam Backup server computer on the corresponding LUN (refer to your SAN documentation).
Direct SAN connection is not available, failing over to network mode..."

Is this a known Problem ? Waht means this Message ?

Thanks for Tips

Michael

[dsmithpdx]

Hi Gostev et. al.,

I am also receiving this "Unable to allocate processing resources" error, but our proxy is configured to use the Virtual Appliance transport mode rather than Direct SAN Access. Will this hotfix resolve our issue as well? We have four separate jobs that run, backing up a total of 17 VMs. All VMs backup successfully, except for the VM that runs Veeam itself as well as the backup proxy. That's the only one reporting this error. Any suggestions?

Thanks,
Doug

[tsightler]

Do you have failover to network mode disabled? A server cannot use hotadd to backup itself (known issue in release notes due to VDDK limitation). The system should detect this and failover to network mode or to another proxy, however, if network mode is disabled, or there is no other proxy available, then the message is technically correct, it is "Unable to allocate processing resource" due to the fact that no appropriate resources are available.

[dsmithpdx]

Indeed I do have failover to network mode disabled. Thank you!

[jlh]

Hello,

it seems everyone is happy now that the patch is here, but me...

I am evaluating B&R 6 and I startet testing SAN access. It didn't work. I found the patch and installed it. For what ever reason after that some VMs worked and others didn't. Today I wanted to dig a little deeper and what worked yesterday, dosen't work today. Acually nothing works anymore. All I had done is turn off the server and back on this morning. Windows can see the LUNs but I'm getting the same errors as before. We have a DELL MD3000i. Any suggestions?

Thanks,
Jerry

[Gostev]

Yes, the patch had helped everyone so far - all the corresponding support cases are closed. First things to check is whether you have all LUNs mounted on your backup proxy server. If this looks good, please contact our support and have them investigate this further for you. Thanks!