Glenn Matthys
Glenn Matthys
I'm unable to keep VerneMQ running with systemd (CentOS 7). `/bin/systemctl start vernemq.service` hangs until it is killed by systemd after 1 hour. Logs: ``` Aug 20 11:35:54 websrv1 vmq[12512]:...
I have vernemq set to log to syslog. Here's the entire log: ``` Aug 20 11:35:37 websrv1 systemd: Starting VerneMQ Server... Aug 20 11:35:41 websrv1 vernemq: config is OK Aug...
In the unit file PIDFile is set: `PIDFile=/run/vernemq/vernemq.pid`. But this PID file is never created. That's probably why systemd is waiting.
The problem seems to be here: ``` do_start() { # Make sure there is not already a node running node_down_check # Warn the user if ulimit is too low check_ulimit...
Added `/usr/lib64/vernemq/bin/vernemq start` to /etc/rc.local like a caveman. It outputs the following: ``` [root@websrv1 vernemq]# /etc/rc.d/rc.local Last login: Thu Aug 20 14:24:33 CEST 2020 on pts/1 !!!! !!!! WARNING: ulimit...
Ok so I narrowed it down to something complete different: nodename in vernemq.conf: ``` ## Name of the Erlang node ## ## Default: [email protected] ## ## Acceptable values: ## -...
That makes perfect sense to me, `equal` and `in` operations (and their not versions) should suffice.
You can just do this in your own cache handler, no need to extend fast-memoize for this.
@ryantxu can you please have a look at this?
See #48