Hello everyone.
We have a recurring situation. Some time ago we noticed that the output of fw ctl conntab | grep '<IP-of-LDAP-server,389' | wc -l kept increasing to a point that it started showing numbers bigger than the number of users. We opened a TAC ticket, tried changing the priority of the problematic server from 1 to a higher number and it didn't work. What worked was deleting the server and readding with lower priority. The problem was observed in multiple gateways that used that server.
Now, more than a year later, the same problem is occurring with the same ldap server. Back then we thought that something got 'stuck' somewhere and readding simply resolved it. But now the fact that it is happening again on gateways with different versions (varying from R81.10 Take 110 to R81.20 Take 41) is pretty confusing. Currently the only potential solution is to do what we did in the previous incident.
I am thinking of opening a TAC ticket again, but have a hunch that they will tell us to do the same thing and won't tell us why this is happening in the first place. And we have the experience of not getting satisfactory answers to 'why' questions as long as the issue at hand is not critical.
So before asking for a blind help from TAC, I feel consulting you guys' experience and knowledge will give me some insight.
Thanks as always.
Cheers!