> There still a leak on Trocla database connections (latest
> gem/module/hiera backend).
>
> I've installed a puppetserver, with only a node connected on; after 5
> days, there is 240 opened connections.
So, as you were changing a lot (new gem, module and switching to new
backend) I'm not sure wh
The Trocla leak seems to be in the trocla-hiera-backend
diff --git a/lib/puppet/functions/trocla_lookup_key.rb
b/lib/puppet/functions/trocla_lookup_key.rb
index d377ec8..f61df46 100644
--- a/lib/puppet/functions/trocla_lookup_key.rb
+++ b/lib/puppet/functions/trocla_lookup_key.rb
@@ -33,6 +33,8
Hi,
There still a leak on Trocla database connections (latest
gem/module/hiera backend).
I've installed a puppetserver, with only a node connected on; after 5
days, there is 240 opened connections.
Best regards,
Le 16/10/2017 à 09:44, Poil a écrit :
Hi,
We have upgrade to Hiera Backend
Hi,
We have upgrade to Hiera Backend v5 (and to the new trocla v5 backend)
There still a leak but it seems to be very very low.
I've also switched in our site.pp all hiera/hiera_hash/hiera_array call
to lookup. We still have some hiera* function call in some module, I'm
asking myself if the l
Hi,
The database connections leak remains after upgrading trocla.
Also after 2 days the catalog computation times start to increase again.
I'm going to install newrelic on a puppetserver and if I see nothing I will try
to analyze via your blog article.
Best regardsLe 4 oct. 2017 10:58, Poil a
Hi,
Thanks !
We had Trocla 0.2.3, and module 0.2.2 I'm upgrading to Trocla 0.3.0 and
latest module code
Best regards,
Le 04/10/2017 à 08:27, Peter Meier a écrit :
We are on hiera v3, we only tuned "max-requests-per-instance: 5000"
because of a databases connection leak with our Trocla lib
> We are on hiera v3, we only tuned "max-requests-per-instance: 5000"
> because of a databases connection leak with our Trocla library.
This is fixed with the trocla module 1.0.1 [1] are you on that version?
best
~pete
[1]
https://github.com/duritong/puppet-trocla/commit/bbedb788a7951e2f69c1c2
Depending on what the trocla library does, it could be leaking objects
to the java layer, in which case tuning the max-requests-per-instance
down would not help. In general, the best way to find leaks like you
are talking about is described here:
https://puppet.com/blog/puppet-server-advanced-memor
Hi,
We have PuppetServer 2.8 on RHEL7.
After some days computation of the catalog become slower and slower; the
load average of the compute nodes increased and the compute goes in timeout.
All our 4 computes nodes have 48 cores, only 10% of each core is used
when the timeout occured.
We ar
