1
I would like to know what steps should I take before rebooting a Linux server to avoid losing the configuration. Example: I made a rule with iptables manually, did not add to rc.local. This won't apply when I log on again.
In this case I should use iptables-save and export the rules. But I would like to cover all possible scenarios.
Are there other steps that I need to think about and include for a production environment server?
Marlon
Posted 2016-08-12T21:45:37.790
Reputation: 127
1Speaking as a sysadmin and reading
for a production enviroment server: 1) First test in on the acceptance server. Then document it in the properplace so your colleagues or backups also know what was done and why. Test thouroughly including a reboot of the QA environment, Then and only then: 2) Following the step documented in point 1 apply to production. This assumes you have the normal setup of DEV, QA and PROD servers. – Hennes – 2016-08-12T23:12:47.2731A very careful time consuming process of examining each application one by one for unsaved changes. The fact that nobody documented these, and created back up files is very scary indeed. You have to somehow read the in-memory config and compare it line for line with the config found in the /etc folder, (hopefully its there). Or at least dump the memory version to the correct file in the etc folder. – cybernard – 2016-08-13T05:21:35.567
@Hennes I understand your point. – Marlon – 2017-01-30T21:29:55.967
@cybernard But I'm seeing the fact thinking in a technical view when the customer knows about your middleware (customer scope) and you have to validate a linux server. (you just have docummented infrastructure scope of the project in production server and very small info about middleware) – Marlon – 2017-01-30T21:31:19.653