FUSION disk full 5-6 days after starting new server

[dlukinski]

Dear Nagios support

We've got Fusion re-spinned 3rd time since November only to get it running out of space in a matter of days

Code: Select all

[root@fikc-nagfsprod01 lib]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/VolGroup-lv_root
                       37G   35G  251M 100% /
tmpfs                 1.9G     0  1.9G   0% /dev/shm
/dev/sda1             477M   95M  357M  22% /boot

[root@fikc-nagfsprod01 scripts]# ./truncate_polled.php
PHP Fatal error:  Call to a member function debugDumpParams() on a non-object in /usr/local/nagiosfusion/html/includes/classes/db.inc.php on line 589

[root@fikc-nagfsprod01 mysql]# du -s * | sort -nr | head -n10
34573088        ibdata1
5120    ib_logfile1
5120    ib_logfile0
988     mysql
228     fusion
4       test
[root@fikc-nagfsprod01 mysql]#

This is NOT OK. This NOT the PRODUCT and the SUPPORT we expected. Please fix this situation, where we are unable to use the Product we've paid for 4 MONTH AGO!

[eloyd]

How did you install Fusion?

[dlukinski]

How did you install Fusion?

From the latest VMWARE teamplate provided by Nagios

[bheden]

@dlukinski,

We are working on a special fix for you. We were able to isolate an edge case where the non-fixing of the giant database occurred - so sorry about that.

In the meantime, while waiting on the fix, are you able to increase the size of the disk on your virtual machine in preparation for the fix that I will PM you shortly?

Thank you for your patience!

[dlukinski]

@dlukinski,

We are working on a special fix for you. We were able to isolate an edge case where the non-fixing of the giant database occurred - so sorry about that.

In the meantime, while waiting on the fix, are you able to increase the size of the disk on your virtual machine in preparation for the fix that I will PM you shortly?

Thank you for your patience!

Because of our environment, i have to be very specific about sizes.

so we now have:
- 2 XI servers (insignificant DEV & BACKUP servers plus 1500+ hosts / 8500+ checks PROD server)
- 1 LOG (quite some number of log checks there)
- 8 CORES (appr. 100 services per server)


What would be the right size for the Nagios Fusion Disk allocation and creating separate backup volume (again what should be the size)?
- keeping logs for 1-3 days

I would much rather re-spin, even so i'm not happy overall with the situation (but not working anyways) VS allocating more space we will not require later on

[bheden]

To be honest, under perfect circumstances the size of the disk necessary for Fusion should only need to be about 20GB, maybe 30GB for 15+ servers. As for backups, I can't think of any scenario that would require more than 5GB.

Of course, this is all based on it working exactly as intended, and it is obvious that this isn't the case. We have identified the issue, which seems to have impacted quite a lot of customers. On behalf of Nagios Enterprises, I'd like to apologize to you and anyone else who this issue has effected.

I think once I send you your personal quick fix, you should be able simply to re-spin that up on it's own and be done. I'm trying not to rush getting it to you, because I want to make sure I've fixed the problem. It should be sometime today, hopefully this morning.

[bheden]

@dlukinski,

I've PMed you the OVA. Please let me know if a) that solved your problems, and b) if we can close this thread.

Thanks!

[dlukinski]

Our problems did not go anywhere.
1 week after re-spinning the product we've got 26 GB of storage consumed

This is a disaster product. We want our money back!

[eloyd]

We've helped a number of customers install, configure, and use Fusion on a long-term basis and none are reporting the kind of problems you are. It's definitely not a disaster product, but there must be something specific to your installation that is causing it to do what it is doing.

Have you considered installing from the downloadable installation media at https://www.nagios.com/downloads/nagios-fusion/linux? Starting with a clean Linux install rather than an OVA file may make a difference for you, since the latest package is going to have the database fix built-in.

[dlukinski]

We've helped a number of customers install, configure, and use Fusion on a long-term basis and none are reporting the kind of problems you are. It's definitely not a disaster product, but there must be something specific to your installation that is causing it to do what it is doing.

Have you considered installing from the downloadable installation media at https://www.nagios.com/downloads/nagios-fusion/linux? Starting with a clean Linux install rather than an OVA file may make a difference for you, since the latest package is going to have the database fix built-in.

We've used OVA files from you to exclude any possible questions about OS since the entire server would be supplied by you.
It worked for XI and LOG. We really wanted to get Fusion to work for a long time. Not sure if bringing the extra variable (OS installed by us) would make things any better. All we know that FUSION DB growing through the roof while we've already 3 different OVA files produced by Nagios Developments. We also know that you cannot fix the DB once it is already large, nor could you help us with requirements for our Fusion to run (vs Fused Servers we openly communicated about).

Again, all we need is for your product to work from your OVA as expected and advertised.