Virtual systems failover after policy installation

[vikupoi_123]

When we are doing policy installations in the particular VS of VSX gateway. That VS failover is happening.. 

[Chris_Atkinson]

Do you observe any corresponding CUL messages?

I would suggest sharing further information (including version/JHF level) with TAC to investigate the issue further.

[Kaspars_Zibarts]

We had a "very busy" VS that was impossible to push during day time / peak hour.

Then we separated fwk and rest of the processes on that VS using affinity and all started working OK

fw ctl affinity -l

....

VS_10: CPU 1 21
VS_10 fwk: CPU 13 14 15 33 34 35

so you can see that 6 cores are used only for fwk workers and two (1 and 21) for all other purposes including policy push handling.

But as mentioned earlier you need to supply more info with logs etc, could be something totally different

[vikupoi_123]

Hi Kaspars ,
Sorry for inconvenience for delay response we have VS1 failover at the time of policy installation . Please find the aoutput of "fw ctl affinity -l" and highlight the issue if have ...

[Expert@GWvsx1:0]# fw ctl affinity -l
eth1-04: CPU 2
eth2-01: CPU 0
eth2-02: CPU 3
eth2-03: CPU 4
Mgmt: CPU 0
Sync: CPU 0
VS_0 fwk: CPU 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
VS_1 fwk: CPU 10 11 12 13 14 15 16 17 18 19
VS_2 fwk: CPU 10 11 12 13 14 15 16 17 18 19
Interface eth1-02: has multi queue enabled
Interface eth1-03: has multi queue enabled

[Chris_Atkinson]

Out of interest is this an R80.20 VSX gateway?

[vikupoi_123]

No it is R77.20 GW

[PhoneBoy · Accepted solution]

R77.20 has been End of Support for quite a while now.
Recommend upgrading to a supported release.