Good morning. I was wondering if anyone has experienced a problem
similar to this. Here is the setup.

2 ClusterXL Active / Standby Clusters. 2 members each. One cluster is
now running R60 HFA05 and the other HFA04. Both clusters are RHEL3.0 for
OS and not SPLAT.

All 4 servers are running on DL360's with 2GB or more of ram and are
running in SMP mode. Over the past few months I've noticed that
whichever cluster member is active in each cluster has a fw ctl pstat
read out for hmem as the examples below. I know there were issues with
Broadcom nics and Clusterxl but we have those disabled and they seem to
be out of this scenario.

Internal Cluster Active member pstat for hmem

Hash kernel memory (hmem) statistics:
  Total memory allocated: 31457280 bytes in 7672 4KB blocks using 8
pools
  Initial memory allocated: 6291456 bytes (Hash memory extended by
25165824 byte
s)
  Memory allocation  limit: 31457280 bytes using 512 pools
  Total memory bytes  used: 24679232   unused:  6778048 (21.55%)   peak:
2463684
8
  Total memory blocks used:     7672   unused:        0 (0%)   peak:
7672
  Allocations: 367577057 alloc, 4819638 failed alloc, 367460490 free


External Cluster Active member pstat for hmem

Hash kernel memory (hmem) statistics:
  Total memory allocated: 31457280 bytes in 7671 4KB blocks using 9
pools
  Initial memory allocated: 6291456 bytes (Hash memory extended by
25165824 bytes)
  Memory allocation  limit: 31457280 bytes using 512 pools
  Total memory bytes  used: 26327976   unused:  5129304 (16.31%)   peak:
27140032
  Total memory blocks used:     7671   unused:        0 (0%)   peak:
7671
  Allocations: 2432183286 alloc, 75845782 failed alloc, 2432006434 free

Both clusters have capacity optimization set to auto. Now I would have
believed that this readout was simply a Checkpoint feature but about a
month back upon noticing these number (actually this was going on for 3
or 4 months), I adjusted the memory parameters to about double in the
Capacity Optimization sceen of the Cluster object. Since that time I
have been getting kernal panics about every 7 to 10 days. Does anybody
have any recommendations for hmem settings or info that this may be some
sort of bug in NGX? I do already have a ticket opened with Checkpoint on
this but figured I would shoot out an email to the list. Let me know if
anyone needs any further info from me and thanks in advance.

Thanks

Jeremy Lieb CCSE-NGX CCSE+NGX
Firewall Administrator
Open Text Corporation
100 Tri-State Int'l Pkwy
Third Floor
Lincolnshire, IL 60069
18472679330  ext 4395
 

=================================================
To set vacation, Out-Of-Office, or away messages,
send an email to [EMAIL PROTECTED]
in the BODY of the email add:
set fw-1-mailinglist nomail
=================================================
To unsubscribe from this mailing list,
please see the instructions at
http://www.checkpoint.com/services/mailing.html
=================================================
If you have any questions on how to change your
subscription options, email
[EMAIL PROTECTED]
=================================================

Reply via email to