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

R80.40 Jumbo HF Take #67 is now our GA take

shlomip
Employee Alumnus
Employee Alumnus
0 9 2,551

Hi,

R80.40 Jumbo Hotfix Take #67 is now our GA take

This take is available for download to all via CPUSE (as recommended) and via sk165456

Thanks,

Release Management Group

9 Comments
Roger_Norton
Participant

Since installing Take 67 I've had various issues with the Solr JAVA processes dumping. Anyone else experienced the same issues?

With Take 48 they were stable.

 

 

 

 

shlomip
Employee Alumnus
Employee Alumnus

@Roger_Norton ,

Not something i heard of or known.

Please open a Support Ticket on this one ( and share the ID if you can )

shlomip
Employee Alumnus
Employee Alumnus

@Roger_Norton ,

If you can also provide the process that dumped (cpm or Solr) this could help

 

 

Roger_Norton
Participant

Solr, the JAVA process -D_solr=TRUE. An example dump file name: solr.16718.tar.gz. 

 

In the associated javacore dump txt file:
1TISIGINFO Dump Event "systhrow" (00040000) Detail "java/lang/OutOfMemoryError" "Java heap space" received

 

In solr.elg I get:

WARN [coreLoadExecutor-4-thread-3] org.apache.solr.update.UpdateLog.update:1000 - Unexpected log entry or corrupt log. Entry=8.033935264473067E-42
- java.lang.ClassCastException: java.lang.Double incompatible with java.util.List

 

The server has 16GB RAM and I've not seen it go above ~8.5GGB  even prior to the service crashing.  The way it crashes is that it used all spare CPU capacity for a few minutes, the drops to 1 core while it creates the dump, stops and a new service starts.

 

shlomip
Employee Alumnus
Employee Alumnus

@Roger_Norton ,

Thanks, We are checking.

Appropriate if you can share the Support Ticket ID once you have it

Roger_Norton
Participant

It's taken a while to resolve but the SMS server has now been stable since Check Point support suggested running the command:

/opt/CPrt-R80.40/scripts/run_maintenance -action=TRANSIENT -zip=0 -logs= -cores=/opt/CPrt-R80.40/log_indexes -out= -category=other-smartlog -date=2020-08-29 -days=0 -oldest=0 -clean

Support recommended a full backup and snapshot prior to running the command, so use with care.

I suspect the command resolved the file corruption reported in solr.elg log, but I need to do more digging to confirm this.

Command options:

- Files are scanned up to the date stated.  

- Log index categories include:

  • Firewallandvpn - Connection
  • Audit – Audit
  • Other – Sessions (ONLY SME)
  • Smartevent – correlated events (ONLY SME)
  • other-smartlog – Sessions (Smartlog ONLY)
  • resources – URLS data
  • files – Files data

 

 

G_W_Albrecht
Legend
Legend

General Availability Take

Take_78 is the latest R80.40 Jumbo Hotfix Accumulator 😎

Roger_Norton
Participant

Indeed it is 😀 And so the quest continues, now that I can.  

With R81 on the horizon ....

G_W_Albrecht
Legend
Legend

Feelin`much better now 😎

Labels