Create a Post
cancel
Showing results for 
Search instead for 
Did you mean: 
Vladimir
Champion
Champion
Jump to solution

ARP table size increase is not surviving the reboot

Has anyone run in to it after the upgrade to R80.20?

0 Kudos
1 Solution

Accepted Solutions
JozkoMrkvicka
Mentor
Mentor

We are using /etc/rc.local file for similar cases. In this case, here is one-time command which will make sure that parameter will survive a reboot:

# echo "echo 4096 > /proc/sys/net/ipv4/neigh/default/gc_thresh3" >> /etc/rc.local

 But be aware that the current threshold level for ARP cache on Gaia OS:

  • In Gaia Clish:

    HostName> show arp table cache-size
  • In Expert mode:

    [Expert@HostName:0]# dbget ip:arp:cache_size

Will still give you original value configured via clish command (not that one set via echo).

So maybe the better way would be to use bash/clish command:

# echo "clish -c 'set arp table cache-size 4096'" >> /etc/rc.local

 

Kind regards,
Jozko Mrkvicka

View solution in original post

14 Replies
HeikoAnkenbrand
Champion Champion
Champion

Hi @Vladimir 

This doesn't work after a reboot:

> set arp table cache-size 4096
> save config

A quick and dirty solution:-)

Add the following to the start script until the problem is solved by Check Point TAC!

echo 4096 > /proc/sys/net/ipv4/neigh/default/gc_thresh3

 

➜ CCSM Elite, CCME, CCTE
Vladimir
Champion
Champion

Thank you @HeikoAnkenbrand !

HeikoAnkenbrand
Champion Champion
Champion

Still from the good old SPLAT times under R65:-)

➜ CCSM Elite, CCME, CCTE
0 Kudos
Vladimir
Champion
Champion

$#!!, now I remember running into it years ago 🙂

0 Kudos
HeikoAnkenbrand
Champion Champion
Champion

Still works today. Shouldn't read the R&D team:-)

➜ CCSM Elite, CCME, CCTE
0 Kudos
Vladimir
Champion
Champion

Heiko, in the context of VSX, would this be a global setting for the entire VSX box, or is there a way to do it for individual VS?

0 Kudos
PhoneBoy
Admin
Admin
Looks like a global setting.
0 Kudos
Vladimir
Champion
Champion

Well, if this is issue still persists to date, perhaps sk43772 should be rewritten, as it is still explicitly states:

 

Gaia Portal / Gaia Clish will override any settings placed in the /etc/sysctl.conf file. Any changes made to this file do not take effect after a reboot.

To configure threshold level for ARP cache on Gaia OS:

Note: Gaia OS accepts maximal threshold level of 16384.

  • In Gaia Portal:

    Go to Network Management section - click on ARP page - go to section ARP Table Settings section - enter the desired value in Maximum Entries field.

  • In Gaia Clish:

    HostName> set arp table cache-size <Number_of_Entries>
    HostName> save config

To check the current threshold level for ARP cache on Gaia OS::

  • In Gaia Clish:

    HostName> show arp table cache-size
  • In Expert mode:

    [Expert@HostName:0]# dbget ip:arp:cache_size

Notes:

  • Once set in the Gaia Portal / Gaia Clish, the settings will survive a reboot.
  • Settings are applies immediately (reboot, restart of any services, policy installation are not required)

The Maximum Entries value in the Gaia Portal corresponds to gc_thresh3 parameter in Linux kernel.

Note: In Gaia OS, the value of gc_thresh1 and the value of gc_thresh2 are automatically determined by the value of gc_thresh3 value:

  • value of gc_thresh1 is 1/8 the value of gc_thresh3
  • value of gc_thresh2 is 1/2 the value of gc_thresh3

 

0 Kudos
JozkoMrkvicka
Mentor
Mentor

We are using /etc/rc.local file for similar cases. In this case, here is one-time command which will make sure that parameter will survive a reboot:

# echo "echo 4096 > /proc/sys/net/ipv4/neigh/default/gc_thresh3" >> /etc/rc.local

 But be aware that the current threshold level for ARP cache on Gaia OS:

  • In Gaia Clish:

    HostName> show arp table cache-size
  • In Expert mode:

    [Expert@HostName:0]# dbget ip:arp:cache_size

Will still give you original value configured via clish command (not that one set via echo).

So maybe the better way would be to use bash/clish command:

# echo "clish -c 'set arp table cache-size 4096'" >> /etc/rc.local

 

Kind regards,
Jozko Mrkvicka
HeikoAnkenbrand
Champion Champion
Champion

Hi @JozkoMrkvicka 

This is a very nice solution:

# echo "clish -c 'set arp table cache-size 4096'" >> /etc/rc.local

 Gives me the idea to more dirty hacksSmiley LOL

➜ CCSM Elite, CCME, CCTE
0 Kudos
Itaiw
Employee
Employee

Hi @Vladimir 

Using the  following clish commands to change the arp table size and save the configuration should survive after reboot (and did survive in my tests)

> set arp table cache-size 4096
> save config

Please open a technical service request to CeckPoint support to investigate the problem on your machine.

0 Kudos
Brian_Deutmeyer
Collaborator

How many arp entries can my table hold if my size is 4096?  I'm unsure if this maps 1 to 1, like one arp entry means 1 spot from 4096 (4096 - 1 = 4095 arp entries left)?

0 Kudos
Timothy_Hall
Champion
Champion

Yes it is a one-to-one mapping, the 4096 corresponds to a variable called gc_thresh3 which specifies the hard maximum of IP to MAC address mappings in the table.  gc_thresh2is typically 1/2 of gc_thresh3 and specifies when more aggressive garbage collection starts in an attempt to free up table entries and avoid a "Neighbour table overflow" condition.  This situation can impact performance and was covered on pages 66-70 of my "Max Power" book.

 

Gateway Performance Optimization R81.20 Course
now available at maxpowerfirewalls.com
Brian_Deutmeyer
Collaborator

Perfect!  I just read the pages (2nd edition) and it was helpful.  Thanks again.

0 Kudos

Leaderboard

Epsum factorial non deposit quid pro quo hic escorol.

Upcoming Events

    CheckMates Events