redis-openshift-example
redis-openshift-example copied to clipboard
redis-server stops around 24 hours
Do you find this problem? I don't know why, redis-server process isn't existed, but the files redis.pid and redis.sock are still existed.
I think is something related with memory. I'm currently investigating this problem with guys from openshift on freenode...
Are there any updates on this issue? I'm very interested in deploying Redis with my application.
No update on this issue, I guess.
@snitramodranoel @bjjay I forgot to update the bug, i solved with a trivial cron job that check if the redis-server is running or not. But anyway, I think the openshift team is already working on a cartridge for redis :)
Greet!
Thanks for the tip. Anyways, there's a poll about a redis cartridge on the OpenShift community web site. Maybe you guys already know about it. Let's spread the word and ask for votes ;)
@razorinc I suppose the issue is due to how OpenShift is implemented with SELinux. I'm not an OpenShift developer, but I bet this will only be solved for real when a redis cartridge is out. That's my two cents.
@snitramodranoel Leonardo, I don't think is SELinux itself, but for sure, when the cartdrige will come out, would be way more easy to use it :) //cc @marekjelen
Still no cartridge it seems. To add more details to @razorinc's answer.
Add cron cartridge to your app using procedure at https://openshift.redhat.com/community/blogs/getting-started-with-cron-jobs-on-openshift
script file in .openshift/cron/hourly/redis-check.sh containing
#!/bin/sh
kill -0 cat ${OPENSHIFT_RUN_DIR}/redis.pid
|| ${OPENSHIFT_DATA_DIR}/redis/bin/redis-server ${OPENSHIFT_DATA_DIR}/redis/bin/redis.conf
Then $ git add .openshift/cron/hourly/redis-check.sh $ git commit -a -m "add redis-server check cron job" $ git push origin
Corrected about script since Openshift fiddled with their enviroment variables see here