Create a Post
cancel
Showing results for 
Search instead for 
Did you mean: 
Chammi_Kumarap1
Contributor

Management server slowness in R80.10

I migrated an FWSM firewall to Checkpoint. The management server is on R80.10 and the gateways are running on R77.30. The FWSM had a large number of rules and objects. Post migration, we have 7000 rules and 5000 objects in the dashboard.

I keep running to a problem where java process hogs up all the available CPU and I'm unable to do anything at this point. The dashboard stops responding and closes. When trying to reconnect, I keep getting an Operation Timeout error. After some time (around 15 minutes), java process consumption eventually goes down and only after that I'm able to re login.

We are in the process of cleaning up the rulebase but can't do that either because of this issue. Troubleshooting becomes a nightmare. The management server runs on a VM with 16GB RAM and 16 CPU cores. The java process consumption goes up as high as 1500%. Tried to get assistance from TAC but they took the easy way out by saying it's a problem with the number of rules and objects. But surely, 16GB RAM and 16 CPU cores should be able to handle this.

Any assistance to sort this out would be much appreciated. The JHF take installed is 35.

43 Replies
Timothy_Hall
Legend Legend
Legend

When the CPU spikes, is it spiking in "wa" (waiting for I/O) space as shown by the top command?  If so your disk path is way too slow.  You have allocated plenty of memory and cores, but for a configuration of that size disk I/O bandwidth is still extremely important.   I have seen performance issues like this time and time again with SMS systems configured in VMWare with generous amounts of RAM and CPU, but after some investigation it turns out they are sharing a disk path with umpteen database VMs and/or numerous other VMs that are constantly beating on the disk I/O path. 

Also was your SMS disk provisioned thick (desirable) or thin?

--
My book "Max Power: Check Point Firewall Performance Optimization"
now available via http://maxpowerfirewalls.com.

Gateway Performance Optimization R81.20 Course
now available at maxpowerfirewalls.com
0 Kudos
Kaspars_Zibarts
Employee Employee
Employee

In our lab testing 16GB ram came out as a bottleneck. We ended up with 64GB. Check RAM usage and especially swap state. In a lot of swap is in use you will have to increase memory. Remember that log indexing is much more process hungry too with R80.10. in case you log on the same server. R80 generally is much more resource demanding for sure.

Timothy_Hall
Legend Legend
Legend

Exactly, lack of RAM can certainly cause heavy "wa" CPU load as well due to paging/swapping.

--
My book "Max Power: Check Point Firewall Performance Optimization"
now available via http://maxpowerfirewalls.com.

Gateway Performance Optimization R81.20 Course
now available at maxpowerfirewalls.com
PhoneBoy
Admin
Admin

R80 management definitely needs more RAM than previous releases.

This is especially true in large environments (either large number of objects, large number of CMAs, or both).

I use 16GB and 4 Cores for my manager in my home lab, which provides more than adequate performance for my small environment.

if it was a bigger environment, I would certainly allocate more of both.

Chammi_Kumarap1
Contributor

It's not the RAM. I'm not using any swap memory and the CPU is used on %us rather than %wa. We can safely assume hardware limitations here and focus on the software side. 

0 Kudos
Kaspars_Zibarts
Employee Employee
Employee

do you mind sharing top output?

0 Kudos
Kaspars_Zibarts
Employee Employee
Employee

and the ps aux | grep <nbr of the java process with high cpu>

0 Kudos
Chammi_Kumarap1
Contributor

top -c output

0 Kudos
Kaspars_Zibarts
Employee Employee
Employee

Looking at the busy one

admin     4767 77.0  8.9 2547428 1461220 ?     Ssl  Sep27 7456:29 /opt/CPshrd-R80/jre_64/bin/java -D_CPM=TRUE -Xaot:forceaot -Xmx1024m -Xms192m -

you will notice that it has set max memory to 1GB and is actually using all of it.

I can see that on our systems (that is MDS) we have 4GB allocated for the same process. Not too sure what allocates initial max size (xmx) - could be preset environment variable or it could be calculated based on your RAM size. Something CP dudes will have to answer.

I personally would allocate as mach RAM as you can for R80 - it will utilise all of it Smiley Happy On our MDS and MLM out of 64GB nearly 40GB are used for caching that certainly helps making it faster even though we "really" use only 25GB

Kaspars_Zibarts
Employee Employee
Employee

Looks like a lot is set here

$CPDIR/conf/CpSetupInfo_resourceProfiles.conf 

Also worth checking SK110753 has an interesting statement "16GB of RAM machine launches with a maximum Java Heap Size equals to 1024 MB which is unsufficient."

Chammi_Kumarap1
Contributor

This is what I see in the rule as per the SK you have suggested.

:rule ("UPGRADE: Large env resources profile without SME"
:filter (
:memory_min (8192)
:memory_max (24800)
:upgrade_setup (TRUE)
:smart_event (FALSE)
:dedicated_log_server (FALSE)
)
:result (
:memory (
:initial_value (
:name (NGM_CPM_MIN_HEAP)
:memory_allocation (192m)
)
:max_value (
:name (NGM_CPM_MAX_HEAP)
:memory_allocation (2048m)
)
)

It seems the limit is set to 2GB. Either I'm looking at the incorrect rule or the solution is only related to MDM environments.

Chammi_Kumarap1
Contributor

[Expert@PDC-CP-MGT:0]# ps aux | grep java

admin     4528  0.0  0.0   1736   508 pts/3    R+   10:22   0:00 grep java

admin     4540  0.0  0.7 1158912 121896 ?      SNsl Sep27   6:15 /opt/CPshrd-R80/jre_64/bin/java -D_solr=TRUE -Xdump:directory=/var/log/dump/usermode -Xdump:heap:events=gpf+user -Xdump:tool:none -Xdump:tool:events=gpf+abort+traceassert+corruptcache,priority=1,range=1..0,exec=javaCompress.sh solr %pid -Xdump:tool:events=systhrow,filter=java/lang/OutOfMemoryError,priority=1,range=1..0,exec=javaCompress.sh solr %pid -Xdump:tool:events=throw,filter=java/lang/OutOfMemoryError,exec=kill -9 %pid -Xaggressive -Xshareclasses:none -Xgc:scvTenureAge=1,noAdaptiveTenure -Xmx128m -Xms128m -Dlog4j.configuration=file:/opt/CPrt-R80/conf/solr.log4j.properties -Dpath=/opt/CPrt-R80/jars/aspectjrt-1.7.0.jar:/opt/CPrt-R80/jars/commons-io-2.3.jar:/opt/CPrt-R80/jars/commons-lang-2.6.jar:/opt/CPrt-R80/jars/cxf-core-3.1.0.jar:/opt/CPrt-R80/jars/cxf-java2ws-plugin-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-bindings-soap-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-bindings-xml-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-databinding-aegis-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-databinding-jaxb-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-frontend-jaxws-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-frontend-simple-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-javascript-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-transports-http-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-transports-http-jetty-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-ws-addr-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-ws-policy-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-wsdl-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-common-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-java2ws-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-validator-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-wsdlto-core-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-wsdlto-databinding-jaxb-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-wsdlto-frontend-jaxws-3.1.0.jar:/opt/CPrt-R80/jars/java_is.jar:/opt/CPrt-R80/jars/java_sic.jar:/opt/CPrt-R80/jars/jaxb-xjc-2.2.11.jar:/opt/CPrt-R80/jars/jetty_assist.jar:/opt/CPrt-R80/jars/stax2-api-3.1.4.jar:/opt/CPrt-R80/jars/woodstox-core-asl-4.4.1.jar:/opt/CPrt-R80/jars/wsdl4j-1.6.3.jar:/opt/CPrt-R80/jars/xmlschema-core-2.2.1.jar:/opt/CPsuite-R80/fw1/cpm-server/jackson-annotations-2.5.0.jar:/opt/CPsuite-R80/fw1/cpm-server/jackson-core-2.5.0.jar:/opt/CPsuite-R80/fw1/cpm-server/jackson-databind-2.5.0.jar: -Dsolr.log=/opt/CPrt-R80/log/solr.log -jar start.jar /opt/CPrt-R80/conf/jetty.xml

admin     4582  0.1  1.6 1508404 268484 ?      SNsl Sep27  11:58 /opt/CPshrd-R80/jre_64/bin/java -D_RFL=TRUE -Xdump:directory=/var/log/dump/usermode -Xdump:heap:events=gpf+user -Xdump:tool:none -Xdump:tool:events=gpf+abort+traceassert+corruptcache,priority=1,range=1..0,exec=javaCompress.sh RFL %pid -Xdump:tool:events=systhrow,filter=java/lang/OutOfMemoryError,priority=1,range=1..0,exec=javaCompress.sh RFL %pid -Xdump:tool:events=throw,filter=java/lang/OutOfMemoryError,exec=kill -9 %pid -Xaggressive -Xshareclasses:none -Xgc:scvTenureAge=1,noAdaptiveTenure -Xmx512m -Xms96m -Dupgrade.cores.count= -Dfile.encoding=UTF-8 -DreportingServer.conf.dir=/opt/CPrt-R80/conf -Dlog4j.configuration=file:/opt/CPrt-R80/conf/rfl.log4j.properties -DReportingServer.log=/opt/CPrt-R80/log -cp /opt/CPrt-R80/jars/* com.checkpoint.core.LogCore -type jms

admin     4635  0.7  4.2 1476108 689780 ?      Ssl  Sep27  71:10 /opt/CPshrd-R80/jre_64/bin/java -D_smartview=TRUE -Xdump:directory=/var/log/dump/usermode -Xdump:heap:events=gpf+user -Xdump:tool:none -Xdump:tool:events=gpf+abort+traceassert+corruptcache,priority=1,range=1..0,exec=javaCompress.sh smartview %pid -Xdump:tool:events=systhrow,filter=java/lang/OutOfMemoryError,priority=1,range=1..0,exec=javaCompress.sh smartview %pid -Xdump:tool:events=throw,filter=java/lang/OutOfMemoryError,exec=kill -9 %pid -Xaggressive -Xshareclasses:none -Xgc:scvTenureAge=1,noAdaptiveTenure -Xmx256m -Xms256m -Djava.io.tmpdir=/opt/CPrt-R80/tmp -Dfile.encoding=UTF-8 -DDedicatedServer=false -DTaskExecThreads=4 -Djavax.net.ssl.trustStore=/opt/CPshrd-R80/jre_32/lib/security/cacerts -Djavax.net.ssl.trustStorePassword=changeit -Djavax.net.ssl.keyStore=/opt/CPshrd-R80/jre_32/lib/security/cacerts -Djavax.net.ssl.keyStorePassword=changeit -Dlog4j.configuration=file:/opt/CPrt-R80/conf/smartview.log4j.properties -DRTDIR=/opt/CPrt-R80 -Dpath=/opt/CPrt-R80/jars/aspectjrt-1.7.0.jar:/opt/CPrt-R80/jars/commons-io-2.3.jar:/opt/CPrt-R80/jars/commons-lang-2.6.jar:/opt/CPrt-R80/jars/cxf-core-3.1.0.jar:/opt/CPrt-R80/jars/cxf-java2ws-plugin-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-bindings-soap-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-bindings-xml-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-databinding-aegis-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-databinding-jaxb-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-frontend-jaxws-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-frontend-simple-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-javascript-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-transports-http-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-transports-http-jetty-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-ws-addr-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-ws-policy-3.1.0.jar:/opt/CPrt-R80/jars/cxf-rt-wsdl-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-common-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-java2ws-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-validator-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-wsdlto-core-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-wsdlto-databinding-jaxb-3.1.0.jar:/opt/CPrt-R80/jars/cxf-tools-wsdlto-frontend-jaxws-3.1.0.jar:/opt/CPrt-R80/jars/java_is.jar:/opt/CPrt-R80/jars/java_sic.jar:/opt/CPrt-R80/jars/jaxb-api-2.2.7.jar:/opt/CPrt-R80/jars/jaxb-core-2.2.7.jar:/opt/CPrt-R80/jars/jaxb-impl-2.2.7.jar:/opt/CPrt-R80/jars/jaxb-xjc-2.2.11.jar:/opt/CPrt-R80/jars/neethi-3.0.3.jar:/opt/CPrt-R80/jars/rfl_sic.jar:/opt/CPrt-R80/jars/smartview-jetty.jar:/opt/CPrt-R80/jars/woodstox-core-asl-4.4.1.jar:/opt/CPrt-R80/jars/wsdl4j-1.6.3.jar:/opt/CPrt-R80/jars/xmlschema-core-2.2.1.jar: -DSTOP.PORT=8079 -DSTOP.KEY=smartview -jar start.jar OPTIONS=Server,resources,websocket /opt/CPrt-R80/conf/smartview-jetty.xml /opt/CPrt-R80/conf/smartview-service-jetty.xml

admin     4767 77.0  8.9 2547428 1461220 ?     Ssl  Sep27 7456:29 /opt/CPshrd-R80/jre_64/bin/java -D_CPM=TRUE -Xaot:forceaot -Xmx1024m -Xms192m -Xgcpolicy:optavgpause -Djava.io.tmpdir=/opt/CPsuite-R80/fw1/tmp -Xaggressive -Xshareclasses:none -Djava.security.krb5.conf=/opt/CPsuite-R80/fw1/conf/krb5.conf -Xjit:exclude={com/checkpoint/management/dleserver/coresvc/internal/SchemaMgrSvcImpl.getClassInfo*},exclude={com/checkpoint/management/object_store/ObjectStoreSessionImpl.findFieldsBySearchQueryEx*} -Xdump:directory=/var/log/dump/usermode -Xdump:heap:events=gpf+user -Xdump:tool:none -Xdump:tool:events=gpf+abort+traceassert+corruptcache,priority=1,range=1..0,exec=javaCompress.sh CPM %pid -Xdump:tool:events=systhrow,filter=java/lang/OutOfMemoryError,priority=1,range=1..0,exec=javaCompress.sh CPM %pid -Xdump:tool:events=throw,filter=java/lang/OutOfMemoryError,priority=1,exec=kill -9 %pid -Dfile.encoding=UTF-8 -cp /opt/CPshrd-R80/jars/solr-solrj-v4_8_1.jar:* com.checkpoint.management.cpm.Cpm -s

admin     5023  0.0  1.9 439680 314756 ?       Ssl  Sep27   5:16 /opt/CPshrd-R80/jre_32/bin/java -Xmx256m -Xms128m -Xshareclasses:none -Dfile.encoding=UTF-8 -Djetty.home=/opt/CPshrd-R80/jetty -Djava.io.tmpdir=/opt/CPsuite-R80/fw1/tmp -Djetty.state=/opt/CPsuite-R80/fw1/api/conf/jetty.state -DSTOP.PORT=8078 -DSTOP.KEY=checkpointkey -Dlog4j.configuration=file:/opt/CPsuite-R80/fw1/api/conf/log4j.properties -Dtdlog.logDir=/opt/CPsuite-R80/fw1/log -Dtdlog.web_api.logFile=api.elg -Dtdlog.output.appender=elgfile -Dtdlog.web_api.csvFile=api.csv -Dtdlog.output.csv.appender=csvfile -Djetty.host=0.0.0.0 -Dpath=/opt/CPsuite-R80/fw1/api/lib/web_api_jetty.jar: -Xdump:directory=/var/log/dump/usermode -Xdump:heap:events=gpf+user -Xdump:system:none -Xdump:system:events=gpf+abort+traceassert+corruptcache -Xdump:tool:none -Xdump:tool:events=gpf+abort+traceassert+corruptcache,priority=1,range=1..0,exec=javaCompress.sh WEB_API %pid -Xdump:tool:events=systhrow,filter=java/lang/OutOfMemoryError,priority=2,range=1..0,exec=javaCompress.sh WEB_API %pid -Xdump:tool:events=throw,filter=java/lang/OutOfMemoryError,priority=1,exec=kill -9 %pid -jar /opt/CPshrd-R80/jetty/start.jar OPTIONS=Server /opt/CPsuite-R80/fw1/api/conf/jetty.xml

admin     7113  1.9  1.5 1770360 250776 ?      Sl   Sep27 188:11 /opt/CPshrd-R80/jre_64/bin/java -D_CPM_SOLR=TRUE -Xmx512m -Xms64m -Xgcpolicy:optavgpause -Djava.io.tmpdir=/opt/CPsuite-R80/fw1/tmp -Xaggressive -Xshareclasses:none -Xdump:heap:events=gpf+user -Xdump:directory=/var/log/dump/usermode -Xdump:tool:none -Xdump:tool:events=gpf+abort+traceassert+corruptcache,priority=1,range=1..0,exec=javaCompress.sh CPM_SOLR %pid -Xdump:tool:events=systhrow,filter=java/lang/OutOfMemoryError,priority=1,range=1..0,exec=javaCompress.sh CPM_SOLR %pid -Xdump:tool:events=throw,filter=java/lang/OutOfMemoryError,priority=1,exec=kill -9 %pid -Dsolr.solr.home=/opt/CPsuite-R80/fw1/Solr/solr/ -DNGM.SOLR.LOG.DIR=/opt/CPsuite-R80/fw1/log -Djava.util.logging.config.file=/opt/CPsuite-R80/fw1/Solr/etc/logging.properties -DSTART=/opt/CPsuite-R80/fw1/Solr/start.config -Djetty.home=/opt/CPsuite-R80/fw1/Solr/ -DSTOP.KEY=checkpointkey -DSTOP.PORT=8982 -Dpath=/opt/CPsuite-R80/fw1/cpm-server/java_is.jar:/opt/CPsuite-R80/fw1/cpm-server/java_sic.jar:/opt/CPshrd-R80/jars/jetty_assist.jar -jar /opt/CPsuite-R80/fw1/Solr/start.jar

[Expert@PDC-CP-MGT:0]#

0 Kudos
Timothy_Hall
Legend Legend
Legend

Uh, yeah 1382% CPU utilization by the CPM process is definitely not normal.  The main function of CPM is to service all the SmartConsole GUI connections and how they interact with the SMS configuration database.  Try the following:

0)  Tell all other administrators to get out of the SmartConsole.  In the SmartConsole go to Manage & Settings...Sessions.  Do you see any active sessions besides your own, if so try discarding/disconnecting them as perhaps CPM is stuck in some kind of loop trying to service a dead or stuck session.

1) Check the log file for the process, in this case $CPDIR/log/cpm.elg (or it might be $FWDIR/log/cpm.elg).  Perhaps it is constantly barfing some kind of error message there that will provide some insight.

2) Any core dumps for this process in /var/log/dump/usermode?  Based on your provided output the process does not appear to be crashing and restarting, so my guess is no in this case.

3) lsof can be used to see the files and/or network connections a process is currently interacting with, in the case of a runaway process this can give an important clue about what resources the process is trying to access.  Run lsof -p 4767 to see all open files held by that process, and lsof -i -a -p 4767 for a list of network sockets in use.

4) Finally, you can install the strace utility and attach it to the running process with the -p option to monitor all system calls the process is making in its runaway state.  This will give you a pretty definitive idea of what the heck is going on with the process, however be warned that installing strace onto Gaia is most definitely not supported.  Here is a statically-linked binary version of strace that should work and avoid any library dependency issues:

http://vault.centos.org/3.8/os/i386/RedHat/RPMS/strace-4.5.14-0.EL3.1.i386.rpm 

--
My book "Max Power: Check Point Firewall Performance Optimization"
now available via http://maxpowerfirewalls.com.

Gateway Performance Optimization R81.20 Course
now available at maxpowerfirewalls.com
0 Kudos
Chammi_Kumarap1
Contributor

There were some sessions shown in the sessions pane but all of them except the current session were in "Disconnected" state. I have discarded all the disconnected sessions as well but there was no improvement.

There were no core dumps in /var/log/dump/usermode. I have taken the lsof output and the cpm.elg files.

CP logs - Google Drive 

0 Kudos
Chammi_Kumarap1
Contributor

Attached the top command output and the ps aux output for all java processes.

I also noticed that I'm not experiencing this slowness on the test setup which I used for the migration. The test environment has only 4 CPU cores. The difference between the production and the test setup is the logs. This may have a connection with the logging process. I have disabled log indexing as well but it didn't do any good.

0 Kudos
Timothy_Hall
Legend Legend
Legend

Definitely some exceptions getting logged into into cpm.elg.  Also some of these other logfiles might be helpful:

/var/log/opt/CPsuite-R80/fw1/log/cpm.stdout.elg
/var/log/opt/CPshrd-R80/cpstart.log
/var/log/opt/CPsuite-R80/fw1/log/install_policy.elg
/var/log/opt/CPsuite-R80/fw1/log/dbsync.elg

I'd suggest working with Ran Kopelman who has posted to this thread, as diagnosing issues such as these will require interaction with Check Point support.  Tough to tell if the entries in cpm.elg are related to your problem or not.

--
My book "Max Power: Check Point Firewall Performance Optimization"
now available via http://maxpowerfirewalls.com.

Gateway Performance Optimization R81.20 Course
now available at maxpowerfirewalls.com
0 Kudos
Ran_Kopelman
Employee
Employee

Hi Chammi 

 

I'm Ran and I'm a manger in the R&D of Check Point, responsible for I/S in the Management Server.

I would like to help,

Can you please upload all cpm.elg log files (cpm.elg, cpm.elg.1.,2....cpm.elg.10) located under $MDS_FWDIR/log.

Dameon, who can assist Chammi with this procedure, basically I just need these files.

Dameon Welch Abernathy

Thanks

Ran

Ran_Kopelman
Employee
Employee

Also, please run :

kill -3 ‘the pid of CPM’

The output is a tgz, it will be under /var/log/usermode/dump.

Please upload it in addition to the log files.

Again, please use the conventional ways to share this data.

Dameon Welch Abernathy

*The pid of CPM can be found by running 'ps -ef | grep CPM'

 

Thanks,

Ran 

0 Kudos
PhoneBoy
Admin
Admin

Chammi Kumarapathirage‌ if you haven’t already done so, please open a TAC ticket: Contact Support | Check Point Software 

Ran, I’ve also provided you Chammi’s direct contact info offline.

Hugo_vd_Kooij
Advisor

If I recall the top output correctly I noticed some code is run by 32 bit JAVA engines. So you can't profit from all that RAM there.

It is worth while to use the 'c' option to toglle to full commandline output in your top output.

<< We make miracles happen while you wait. The impossible jobs take just a wee bit longer. >>
Jason_Carrillo
Collaborator

I'm getting something similar immediately after upgrading. 400-500% CPU usage only on my Domain Log Manager though, not the Domain Management Server. 

Support is telling me this is expected behavior, that java, on behalf of the solr processing, will use up CPU to the maximum, but will kindly relinquish control when other processes require CPU.

¯\_(ツ)_/¯

0 Kudos
Kaspars_Zibarts
Employee Employee
Employee

yeah, regarding the log server RAM and CPU - we had to more than quadruple it from 4 cores 16GB to 16 cores 128GB to get reasonable performance.. Otherwise logs were unusable

0 Kudos
Timothy_Hall
Legend Legend
Legend

Correct, this heavy CPU usage by SOLR is for log indexing.  If you look closely in ps/top you'll see that the process has been "nice'd" down to the minimum CPU priority possible.  So if literally any other process on the SMS is looking for CPU slices, SOLR will get thrown off the CPU to make way for the other process.

--
My Book "Max Power: Check Point Firewall Performance Optimization"
Second Edition Coming Soon

Gateway Performance Optimization R81.20 Course
now available at maxpowerfirewalls.com
venkata_marutur
Contributor

Hey Chammi, I was in a similar situation like yours. I can for sure say that your hardware specs are not sufficient considering number of rules and objects you have and yes R80+ will require more memory/CPU comparatively.

My suggestion:

- Run #cpstat -f log_server mg and #cpstat -f indexer ls on your SMS and see how many logs you are receiving. In my case I was logging tons of dns logs and all of them are being indexed, I turned off logging on some useless rules and that made things little stable.

- Figure out if the specs are sufficient for the logs that you are receiving (assuming your SMS is doing Policy mgmt and Logging)

Hope it helps.

Thanks.

Chammi_Kumarap1
Contributor

It wasn't a hardware issue in the end. We increased RAM upto 64GB but issue remained the same. Checkpoint released a hotfix to address the issue. I will share the particulars later.

Nico_V
Participant

Hello, same issue here. We are testing out the management server in a vm. We started with way too low resources and bumped up the ram and cpu, still slow.
I wonder if the java processes' heap size is decided during the installation, and future changes to the ram are not taken into account... (unfounded assumption, of course).

I am curious to hear what the hotfix did to address this.

Regards

0 Kudos
Chammi_Kumarap1
Contributor

We asked for particulars regarding the hotfix but didn't get a proper response.The hotfix was supposed to improve UI response. After installing it, we didn't notice any reduction in CPU utilization but we did notice a huge improvement in the UI response time.

SR number if anyone is interested in more information.

1-9779162041

Timothy_Hall
Legend Legend
Legend

Hi Nico,

SMS Java heap sizes are determined every time the system is booted in response to more memory or other resources being added, see my lengthy exploration of this subject here:

https://community.checkpoint.com/message/17639-re-mangement-server-r8010-slowness?commentID=17639?sr... 

--
Second Edition of my "Max Power" Firewall Book
Now Available at http://www.maxpowerfirewalls.com

Gateway Performance Optimization R81.20 Course
now available at maxpowerfirewalls.com
Nico_V
Participant

Thank you both Smiley Happy

0 Kudos

Leaderboard

Epsum factorial non deposit quid pro quo hic escorol.

Upcoming Events

    CheckMates Events