Jump to content


dyg

Slow OSD with only one Distribution Point (Virtual Machine)

Recommended Posts

Hi,

 

In my environment I have 3 locations for Operation System Deployments. Each one has its own DP and boundaries group.

We have a very slow OSD process in one of them (3-4 for one deployment, other DPs take less than an hour)

The only difference betwen the slow DP and the others DPs is that it's a VM with WS 2012. The other two DPs are WS 2008 and they are physical servers.

 

We tried to change the NIC on VM Server with no luck.

We also checked possible networking issues with the infraestructure team and we didn't notice any problem.

 

Do you guys think that it could be a VM problem?

 

Any help would be appreaciated

 

Thanks and regards

 

Share this post


Link to post
Share on other sites


The #1 killer for VM performance is going to be disk IO. I'm willing to bet you VM is sharing it's disk with other VMs. Talk to the VM team and get them to look at thing.

Share this post


Link to post
Share on other sites

Agreed, I was in the same situation and have them move the disk to a dedicated section and it sped things up a lot.

Share this post


Link to post
Share on other sites

Hi,

 

I'm sorry for the delay. We moved the disk to a dedicated sections with no changes.

 

What else could it be?

Share this post


Link to post
Share on other sites

Network? What exactly did you troubleshoot? Did changing the disk change anything? What exactly does moved the disk to a dedicated section mean?

Share this post


Link to post
Share on other sites

The networking team monitored the OSD process and saw no packet loss. We also tried to move a file directly from the server to a client and the transfer was with a normal time.

As you suggested, we have isolated the VM disk to discard IO problems

It seems to be the SCCM OSD process that creates a bottleneck or something else.

Share this post


Link to post
Share on other sites

So if it work fine two other DPs, however one is not working correctly, how exactly is the OSD process creating a bottleneck? Have you compared Perfmon between a working DP and non-working DP?

So you have put the DP on dedicated Disks?

Just because there is no packet lost, doesn't mean the network is not still the problem.

Share this post


Link to post
Share on other sites

I think the same, with other DP the OSD process works well but I have reviewed the server settings and can not find the reason. I do not know if I have to look at any other configuration on this server.

I have compared the perfomance but I have not found high CPU, RAM, network, IO consumptions that can cause the problem.

Correct, the DP is on a dedicated disk.

I do not understand because copying a file manually works fine but via OSD no. I have checked that during the Task sequence point to the correct DP and it is not pointing to another site.

Other servers in the same VM pool do not have similar problems

It's disconcerting.

Share this post


Link to post
Share on other sites

So you have more digging to do but IMO the problem is with the VM or network. Again, if you have two DPs that are working great then, the problem is the extra DP.

 

Have you confirm that your boundaries are correct? Have you confirm that BITS is not a problem?

Share this post


Link to post
Share on other sites

Yes, the boundaries are correct

 

What do you mean that it could be a BITS problem? What would be the correct configuration? I have reviewed the inetpub logs and I didn't see errors. I also did the following test to verify that it was ok.

 

https://blogs.technet.microsoft.com/carlossantiago/2009/04/28/how-to-check-if-a-bits-enabled-distribution-point-is-up-and-running/

 

The only difference with the other DPs is that this is VM and WS2012. The rest are physical and WS2008

Share this post


Link to post
Share on other sites

 

Yes, the boundaries are correct

 

What do you mean that it could be a BITS problem? What would be the correct configuration? I have reviewed the inetpub logs and I didn't see errors. I also did the following test to verify that it was ok.

 

https://blogs.technet.microsoft.com/carlossantiago/2009/04/28/how-to-check-if-a-bits-enabled-distribution-point-is-up-and-running/

 

The only difference with the other DPs is that this is VM and WS2012. The rest are physical and WS2008

 

There is no silver bullet for this, you will need to dig deeper into the VM to determine why it is slower than the physical servers. The issue will not be directly related to CM.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...