Storage mountpoint for new Nagios XI system
-
meganoliver
- Posts: 6
- Joined: Wed Jan 09, 2019 11:26 am
Storage mountpoint for new Nagios XI system
For a new Nagios XI system design, what is the recommended storage mountpoint? Should the storage be partitioned? Could symlinks be used?
Re: Storage mountpoint for new Nagios XI system
Hi @meganoliver,
Shamas shared your previous e-mail with us. Going through those points:
- /var/lib/mysql and /usr/local are both good mountpoints. Together, those should cover the vast majority of data that Nagios XI writes to file.
- We typically don't tell people how to size their disks, because disk usage can vary by several orders of magnitude depending on what you monitor. For instance, A 100-port switch will contain bandwidth data for each port, each of which takes as much space as a server that just needs to be pinged every few minutes. Your best bet is to monitor a couple of devices on a test instance, so that you know how they behave when 'OK' and 'non-OK', and to extrapolate based on that.
- Symlinks should be fine to use, as long as the internal structure of /usr/local/nagios and /usr/local/nagiosxi are preserved, and they stay at the same paths relative to each other.
- We typically don't recommend that people use SAN to store data for Nagios XI, since XI tends to be fairly resource-intensive and you want the fastest read/write possible. However, it sounds like you're trying to allocate resources that are more performant than the regular disk, so that's up to you.
- There's also the issue of what happens if the storage becomes disconnected for any reason. The application wasn't originally designed for systems where directories can disappear, so if that happens you'll need your own solution in place to handle it (or accept that data will be lost for the duration of that outage).
Hopefully this helps, please let us know if you have any other questions for concerns.
Shamas shared your previous e-mail with us. Going through those points:
- /var/lib/mysql and /usr/local are both good mountpoints. Together, those should cover the vast majority of data that Nagios XI writes to file.
- We typically don't tell people how to size their disks, because disk usage can vary by several orders of magnitude depending on what you monitor. For instance, A 100-port switch will contain bandwidth data for each port, each of which takes as much space as a server that just needs to be pinged every few minutes. Your best bet is to monitor a couple of devices on a test instance, so that you know how they behave when 'OK' and 'non-OK', and to extrapolate based on that.
- Symlinks should be fine to use, as long as the internal structure of /usr/local/nagios and /usr/local/nagiosxi are preserved, and they stay at the same paths relative to each other.
- We typically don't recommend that people use SAN to store data for Nagios XI, since XI tends to be fairly resource-intensive and you want the fastest read/write possible. However, it sounds like you're trying to allocate resources that are more performant than the regular disk, so that's up to you.
- There's also the issue of what happens if the storage becomes disconnected for any reason. The application wasn't originally designed for systems where directories can disappear, so if that happens you'll need your own solution in place to handle it (or accept that data will be lost for the duration of that outage).
Hopefully this helps, please let us know if you have any other questions for concerns.
As of May 25th, 2018, all communications with Nagios Enterprises and its employees are covered under our new Privacy Policy