shinken
shinken copied to clipboard
naparuba
All hosts and services are in pending state under Ubuntu Xenial
Hello.
I had a working installation of shinken in a Ubuntu Trusty (14.04 LTS) server. The server died and I reinstalled shinken in a new Ubuntu Xenial (16.04 LTS) server.
All my hosts and services are in pending state.
I can see in the broker log file many communication errors such as these ones:
INFO: [broker-master] Connection problem to the poller poller-master: Connection error to http://localhost:7771/ : Operation timed out after 3001 milliseconds with 0 bytes received
WARNING: [broker-master] Connection problem to the poller poller-master: Connection error to http://localhost:7771/ : Operation timed out after 120001 milliseconds with 0 bytes received
INFO: [broker-master] Connection OK to the poller poller-master
I tried to make a HTTP request to http://localhost:7771/ with wget and it works. I tried to disable the firewall but it did not change anything.
I saw issue #1659 and checked the pycurl version but I am already at the latest version (7.43.0). I also tried to increase thread pool daemons but it did not change anything.
Note that I restarted multiple times all the shinken daemons and shinken could get hosts and services states once. But this did not last, states were not updated. And after a restart, all states were in pending state again.
Did somebody encounter this issue? Did I misconfigured something?
Thanks a lot.
Hi @TiJof.
After creating this issue, I discovered #1885 which reports the same issue on Debian.
You can have a look at it, we discussed about a few tips to try to get it work but we did not get any satisfying result...
Can't find any rpm to setup shinken ... So, I manage to install with pip. The comportement is the same as in Debian 8.7.
Really ??? Shinken RPMs are on Fedora EPEL (https://admin.fedoraproject.org/pkgdb/package/rpms/shinken/).
My bad ... I've only test with centos repo (seems quite normal to not search into fedora, no ?) Anyway, will try asap with the fedora's one (fresh reinstall in progress)
On CentOS you must install EPEL with this command -> yum install epel-repo And install shinken with yum install shinken shinken-arbiter, etc...
Has there been any work at all on this? I have the same problem on two different systems.
And no moving to centos and RPM is not an option.
It sounds like nothing changed in this ticket since it was created.
On my side I now use Naemon and Thruk as web interface. It works fine.
During the migration, I was able reuse a big part of my Shinken configuration since both Shinken and Naemon are compatible with Nagios.
I had the same problem, i changed for Naemon, quite easy to migrate (same Nagios core). Still using Nconf and Thruk with Naemon
