rhuhman1 Absent Member.
Absent Member.
3462 views

Frequent high CPU normal?

I have been working on a ZCM 10.3.1 issue for the last two days which are causing long login times as well as high CPU usage (94 - 100%) I implimented TID 7006967 with some success. So far I have had better login times but have noticed that the CPU report on my VMWare server is still showing frequent high utilization (100%). I am noticing it stays high for a minute may be 2 and then drops to what I would have expected to be normal levels. This pattern is constant almost looking like a heart beat. From what I could see the login times are still normal but I am concerned that once I have students trying to log back in tomorrow morning that the work load will cause the CPU's to go back to high utilization all the time causing long login times. The login times were between 15 and 30 minutes. All during this time the CPU utilization would stay around 94% to 100% (VMWare performance graph).

Is the CPU usage I am seeing now normal or should I expect more issues as the number of computers start to login to the network?

SLES10.2 on VMWare ESX3.5
ZCM10.3.1 with TID 7006967 implemented.
(Tried to install ZRS with a failure that I stopped working on once this issue started Monday morning. The ZRS install failure was on Wednesday and again on Thursday. Didn't think it was related but thought I would mention it.)

Thanks

Rich
Labels (2)
0 Likes
11 Replies
Micro Focus Expert
Micro Focus Expert

Re: Frequent high CPU normal?

For Schools, I recommend implementing the Filters referenced here:
http://www.novell.com/support/viewContent.do?externalId=7004256

School Environments tend to see this issue the most.

If you tell me the number of Students and Devices as well as your DB
Size, I will have a general idea how much of an issue you might be
seeing here.

(If the DB has grown, too large, Rebuilding the DB after cleaning would
really hlep too....
http://www.novell.com/documentation/zcm10/zcm10_system_admin/?page=/documentation/zcm10/zcm10_system_admin/data/bo119lf.html)


Furthermore, If you are seeing High DB Utilization during the evenings
when nobody is working, it is likely caused by a Migrated ZEN7 Bundle.
Did you migrate old ZDM Bundles?

I would consider remigrating Bundles with the greatest number of
"Actions" as MSIs. The Simple existence of some of those bundles can
cause high DB utilization. The number of "Actions" can only be
determined by running the script from the TID and not just looking at
the bundle in the ZCC.




On 11/2/2010 9:36 PM, rhuhman wrote:
>
> I have been working on a ZCM 10.3.1 issue for the last two days which
> are causing long login times as well as high CPU usage (94 - 100%) I
> implimented TID 7006967 with some success. So far I have had better
> login times but have noticed that the CPU report on my VMWare server is
> still showing frequent high utilization (100%). I am noticing it stays
> high for a minute may be 2 and then drops to what I would have expected
> to be normal levels. This pattern is constant almost looking like a
> heart beat. From what I could see the login times are still normal but I
> am concerned that once I have students trying to log back in tomorrow
> morning that the work load will cause the CPU's to go back to high
> utilization all the time causing long login times. The login times were
> between 15 and 30 minutes. All during this time the CPU utilization
> would stay around 94% to 100% (VMWare performance graph).
>
> Is the CPU usage I am seeing now normal or should I expect more issues
> as the number of computers start to login to the network?
>
> SLES10.2 on VMWare ESX3.5
> ZCM10.3.1 with TID 7006967 implemented.
> (Tried to install ZRS with a failure that I stopped working on once
> this issue started Monday morning. The ZRS install failure was on
> Wednesday and again on Thursday. Didn't think it was related but thought
> I would mention it.)
>
> Thanks
>
> Rich
>
>



--
Craig Wilson - MCNE, MCSE, CCNA
Novell Knowledge Partner

Novell does not officially monitor these forums.

Suggestions/Opinions/Statements made by me are solely my own.
These thoughts may not be shared by either Novell or any rational human.
0 Likes
rhuhman1 Absent Member.
Absent Member.

Re: Frequent high CPU normal?

I am running the internal (OEM) sybase database. The database is about 7.5Gig. I am currently running 832 devices. My current user count is around 900. I don't have any migrated ZEN7 bundles. I decided to build the server fresh. I will start working on the clean up and rebuild of the database.

Thanks

Rich
0 Likes
Micro Focus Expert
Micro Focus Expert

Re: Frequent high CPU normal?

The DB is definitely large for your Size.
I think the TID regarding filters will help.

After running the cleanup process, I would recommend Rebuilding the DB
Twice in a row.

Why Twice?
Not really sure WHY, but I have seen the second rebuild reclaim
additional space that was not reclaimed the first time.

Note: Do not be surprised if your DB sees a 90% reduction in size.

In Short, System Requirement results are stored in the DB for each
bundle for each Device-User combination. So as students move from
device to device, the number of records can grow very very large.

In Business environments where generally there is a 1-1 ratio, it is
much less of an issue.

In Exchange for being able to see the System Requirement Results of all
bundles on all devices directly in the ZCC, you will gain a 90%
reduction in the size of your database.

For one very large school system, They reduced their DB from about 80GB
to under 10GB, simply by clearing out and filtering that data.

In your case, I would not be surprised in the least to see your DB dip
below 1GB in size, especially if you are not using Patch Management.


On 11/3/2010 10:06 AM, rhuhman wrote:
>
> I am running the internal (OEM) sybase database. The database is about
> 7.5Gig. I am currently running 832 devices. My current user count is
> around 900. I don't have any migrated ZEN7 bundles. I decided to build
> the server fresh. I will start working on the clean up and rebuild of
> the database.
>
> Thanks
>
> Rich
>
>



--
Craig Wilson - MCNE, MCSE, CCNA
Novell Knowledge Partner

Novell does not officially monitor these forums.

Suggestions/Opinions/Statements made by me are solely my own.
These thoughts may not be shared by either Novell or any rational human.
0 Likes
rhuhman1 Absent Member.
Absent Member.

Re: Frequent high CPU normal?

Is this something I should place on my monthly Maintenance schedule to help keep the size down. We are a one to one laptop environement but still tend to have a lot of the end users logging into other computers then there own. I am running ZPM. Do I need to perform any further maintenance to the database because of this or will the clean up and rebuild help with it? Do I need to look at clearing out any other files when I run this maintenance? Say log files or unneeded patch files that ZCM isn't going to do itself?

Thanks for the help. I didn't run the TID you sent since it talked about 10.2 and we are on 10.3.1

Thanks

Rich
0 Likes
Micro Focus Expert
Micro Focus Expert

Re: Frequent high CPU normal?

No, this does not need to be done often.

However, in your case your tables have grown very large because of this
specific data they were recording. The filters will prevent this
specific data from being saved (and most ppl don't even use or pay
attention to this data much.)

However, the tables/indexes will need to be rebuilt again after the DB
is reduced in size by 80-90% to become highly efficient.

Maybe an Annual thing, but definitely nothing that regular.

On 11/3/2010 11:36 AM, rhuhman wrote:
>
> Is this something I should place on my monthly Maintenance schedule to
> help keep the size down. We are a one to one laptop environement but
> still tend to have a lot of the end users logging into other computers
> then there own. I am running ZPM. Do I need to perform any further
> maintenance to the database because of this or will the clean up and
> rebuild help with it? Do I need to look at clearing out any other files
> when I run this maintenance? Say log files or unneeded patch files that
> ZCM isn't going to do itself?
>
> Thanks for the help. I didn't run the TID you sent since it talked
> about 10.2 and we are on 10.3.1
>
> Thanks
>
> Rich
>
>



--
Craig Wilson - MCNE, MCSE, CCNA
Novell Knowledge Partner

Novell does not officially monitor these forums.

Suggestions/Opinions/Statements made by me are solely my own.
These thoughts may not be shared by either Novell or any rational human.
0 Likes
rhuhman1 Absent Member.
Absent Member.

Re: Frequent high CPU normal?

Craig,
Thanks for the help so far. Quick question. When I run the rebuild steps I am getting some sort of SQL Authentication error during the dbunload portion. I am assuming that this is from the step 8f not being correct. When it states to place the results of the output_filename are they just wanting to Alphanumeric code at the end of that line? or did I get that section incorrect?

When I completed the steps and looked at the db file it was only 4.9mb which I didn't think would be correct since the starting size was 7.6. I did try to move that over and start the database but just confirmed that it was a non usable database.

Thanks for the help.

Rich
0 Likes
Micro Focus Expert
Micro Focus Expert

Re: Frequent high CPU normal?

Send an email to craig_d_wilson at yahoo, and I will email you a
pre-edited file you can use and some details.

Mention if you have Windows or Linux, just so I can be clear if necessary.

I'm not sure what would be 4.9mb, If you mean GB, that could be correct.

After rebuilding the DB twice, I would not be surprised to see a 80-90%
decrease in size.

On 11/3/2010 6:06 PM, rhuhman wrote:
>
> Craig,
> Thanks for the help so far. Quick question. When I run the rebuild
> steps I am getting some sort of SQL Authentication error during the
> dbunload portion. I am assuming that this is from the step 8f not being
> correct. When it states to place the results of the output_filename are
> they just wanting to Alphanumeric code at the end of that line? or did I
> get that section incorrect?
>
> When I completed the steps and looked at the db file it was only 4.9mb
> which I didn't think would be correct since the starting size was 7.6.
> I did try to move that over and start the database but just confirmed
> that it was a non usable database.
>
> Thanks for the help.
>
> Rich
>
>



--
Craig Wilson - MCNE, MCSE, CCNA
Novell Knowledge Partner

Novell does not officially monitor these forums.

Suggestions/Opinions/Statements made by me are solely my own.
These thoughts may not be shared by either Novell or any rational human.
0 Likes
rhuhman1 Absent Member.
Absent Member.

Re: Frequent high CPU normal?

Craig,

I sent the e-mail last night.

Thanks

Rich
0 Likes
danagosto2 Absent Member.
Absent Member.

Re: Frequent high CPU normal?

I am having a fairly similar issue that we've been trying to figure out for the last few months.

What we did see originally was before staff/students came back to school everything was fine, but immediately, our primary server just choked bad. Ended up finding out that the auto refresh for the launcher was set to 15 minutes and after switching to manual we didn't have any big problems.

Now, we end up rebooting our two primary servers pretty much daily since the login times for Win7 is slowing quite rapidly and responsiveness of imaging and the ZCM web interface gets really slow. The biggest thing I see is that the Free memory drops 6.5Gb's within an hour (down to 0) on both primary servers when our district logs into the computers.

Background on our current setup:
1 - Sattelite authentication server, 56Gb Ram, 8x2.4Ghz Zeon's, Win Serv 2008 Ent 64bit on a physical HP DL585 G5 (I know it's overkill, was just a spare server)
2 - Primary Servers (using closest server rule to divide the district in half), Running Win Serv 2008 Std 64bit, 8Gb Ram, 4x 3.2Ghz vCpu's, running inside VMware ESX 4.0u1
1 - Database server, dedicated to ZCM only, Win Serv 2008 Std 64bit, 8Gb Ram, 4x3.2Ghz vCpu's, running inside VMware ESX 4.0u1, SQL 2008

Our first primary server is also the preboot server. Imaging is done off of either one of the primary servers, but not during peak user hours (we're working on getting another server built to dedicate to imaging and one for management). And our management is done off of either server as well and we only have 15 people that use the web interface of ZCM.

We have 4,250 devices and about 15,000 users running Windows 7 32bit.

ZCM version is 10.3.0 and will be updated by our Zen guy to 10.3.1 by next Friday.

The database we have is currently 32.5Gb and teh zStatusEvent table is 4.2Gb and zObjectInfo table is 7.6Gb.

I'm wondering if applying the 7004256 TID will do anything for us and what effects removing the "FullSysReqEvaluation" status's out of that table will do for us.

I think I have enough info in here, but if you need more, let me know. Thanks in advance for anyone that can help my sanity!
0 Likes
danagosto2 Absent Member.
Absent Member.

Re: Frequent high CPU normal?

I am having a fairly similar issue that we've been trying to figure out for the last few months.

What we did see originally was before staff/students came back to school everything was fine, but immediately, our primary server just choked bad. Ended up finding out that the auto refresh for the launcher was set to 15 minutes and after switching to manual we didn't have any big problems.

Now, we end up rebooting our two primary servers pretty much daily since the login times for Win7 is slowing quite rapidly and responsiveness of imaging and the ZCM web interface gets really slow. The biggest thing I see is that the Free memory drops 6.5Gb's within an hour (down to 0) on both primary servers when our district logs into the computers.

Background on our current setup:
1 - Sattelite authentication server, 56Gb Ram, 8x2.4Ghz Zeon's, Win Serv 2008 Ent 64bit on a physical HP DL585 G5 (I know it's overkill, was just a spare server)
2 - Primary Servers (using closest server rule to divide the district in half), Running Win Serv 2008 Std 64bit, 8Gb Ram, 4x 3.2Ghz vCpu's, running inside VMware ESX 4.0u1
1 - Database server, dedicated to ZCM only, Win Serv 2008 Std 64bit, 8Gb Ram, 4x3.2Ghz vCpu's, running inside VMware ESX 4.0u1, SQL 2008

Our first primary server is also the preboot server. Imaging is done off of either one of the primary servers, but not during peak user hours (we're working on getting another server built to dedicate to imaging and one for management). And our management is done off of either server as well and we only have 15 people that use the web interface of ZCM.

We have 4,250 devices and about 15,000 users running Windows 7 32bit.

ZCM version is 10.3.0 and will be updated by our Zen guy to 10.3.1 by next Friday.

The database we have is currently 32.5Gb and teh zStatusEvent table is 4.2Gb and zObjectInfo table is 7.6Gb.

I'm wondering if applying the 7004256 TID will do anything for us and what effects removing the "FullSysReqEvaluation" status's out of that table will do for us.

I think I have enough info in here, but if you need more, let me know. Thanks in advance for anyone that can help my sanity!
0 Likes
Micro Focus Expert
Micro Focus Expert

Re: Frequent high CPU normal?

Great Info, but I would recommend a new thread.
It's far easier to follow tips and suggestions.

On 11/5/2010 12:06 PM, danagosto2 wrote:
>
> I am having a fairly similar issue that we've been trying to figure out
> for the last few months.
>
> What we did see originally was before staff/students came back to
> school everything was fine, but immediately, our primary server just
> choked bad. Ended up finding out that the auto refresh for the launcher
> was set to 15 minutes and after switching to manual we didn't have any
> big problems.
>
> Now, we end up rebooting our two primary servers pretty much daily
> since the login times for Win7 is slowing quite rapidly and
> responsiveness of imaging and the ZCM web interface gets really slow.
> The biggest thing I see is that the Free memory drops 6.5Gb's within an
> hour (down to 0) on both primary servers when our district logs into the
> computers.
>
> Background on our current setup:
> 1 - Sattelite authentication server, 56Gb Ram, 8x2.4Ghz Zeon's, Win
> Serv 2008 Ent 64bit on a physical HP DL585 G5 (I know it's overkill, was
> just a spare server)
> 2 - Primary Servers (using closest server rule to divide the district
> in half), Running Win Serv 2008 Std 64bit, 8Gb Ram, 4x 3.2Ghz vCpu's,
> running inside VMware ESX 4.0u1
> 1 - Database server, dedicated to ZCM only, Win Serv 2008 Std 64bit,
> 8Gb Ram, 4x3.2Ghz vCpu's, running inside VMware ESX 4.0u1, SQL 2008
>
> Our first primary server is also the preboot server. Imaging is done
> off of either one of the primary servers, but not during peak user hours
> (we're working on getting another server built to dedicate to imaging
> and one for management). And our management is done off of either
> server as well and we only have 15 people that use the web interface of
> ZCM.
>
> We have 4,250 devices and about 15,000 users running Windows 7 32bit.
>
> ZCM version is 10.3.0 and will be updated by our Zen guy to 10.3.1 by
> next Friday.
>
> The database we have is currently 32.5Gb and teh zStatusEvent table is
> 4.2Gb and zObjectInfo table is 7.6Gb.
>
> I'm wondering if applying the 7004256 TID will do anything for us and
> what effects removing the "FullSysReqEvaluation" status's out of that
> table will do for us.
>
> I think I have enough info in here, but if you need more, let me know.
> Thanks in advance for anyone that can help my sanity!
>
>



--
Craig Wilson - MCNE, MCSE, CCNA
Novell Knowledge Partner

Novell does not officially monitor these forums.

Suggestions/Opinions/Statements made by me are solely my own.
These thoughts may not be shared by either Novell or any rational human.
0 Likes
The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.