set ulimit for server according to redhat spec

[DaanHoogland]

Description

This PR fixes #9872

Types of changes

• Breaking change (fix or feature that would cause existing functionality to change)
• New feature (non-breaking change which adds functionality)
• Bug fix (non-breaking change which fixes an issue)
• Enhancement (improves an existing feature and functionality)
• Cleanup (Code refactoring and cleanup, that may add test cases)
• build/CI
• test (unit or integration test code)

Feature/Enhancement Scale or Bug Severity

Feature/Enhancement Scale

• Major
• Minor

Bug Severity

• BLOCKER
• Critical
• Major
• Minor
• Trivial

Screenshots (if appropriate):

How Has This Been Tested?

How did you try to break this feature and the system with this change?

Also see original ticket, set the contents of /etc/systemd/system/cloudstack-management.service.d/filelimit.conf

# vi /etc/systemd/system/cloudstack-management.service.d/filelimit.conf 
# systemctl daemon-reload
# systemctl restart cloudstack-management
# cat /proc/$(ps aux |grep cloudstack-management|grep -v grep |awk '{print $2}')/limits

[weizhouapache]

@DaanHoogland

• is there some changes for debian ?

• I checked my env

# cat /proc/$(ps aux |grep cloudstack-management|grep -v grep |awk '{print $2}')/limits
Limit                     Soft Limit           Hard Limit           Units     
...
Max open files            524288               524288               files     
...

after applying the changes, run systemctl daemon-reload and restarted mgmt server,

# cat /proc/$(ps aux |grep cloudstack-management|grep -v grep |awk '{print $2}')/limits
Limit                     Soft Limit           Hard Limit           Units     
...
Max open files            500000               500000               files     
...

it works. but what should be the best value ?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 15.13%. Comparing base (47f6019) to head (8c1ef25).
Report is 10 commits behind head on 4.19.

Additional details and impacted files

@@            Coverage Diff            @@
##               4.19   #10040   +/-   ##
=========================================
  Coverage     15.13%   15.13%           
- Complexity    11261    11268    +7     
=========================================
  Files          5408     5408           
  Lines        473842   473890   +48     
  Branches      57771    57787   +16     
=========================================
+ Hits          71696    71704    +8     
- Misses       394145   394182   +37     
- Partials       8001     8004    +3     

Flag	Coverage Δ
uitests	4.30% <ø> (ø)
unittests	15.85% <ø> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[DaanHoogland]

@weizhouapache I think the best value is installation dependant. If it works on debian-likes, there is nothing to do there. This was reported by a user on centos7 and does not pertain to the default value but to the mechs of changing it that didn't work. We can change the value before merging to contain the default. I hadn't looked at that yet. It was just a guestimate.

[weizhouapache]

@weizhouapache I think the best value is installation dependant. If it works on debian-likes, there is nothing to do there. This was reported by a user on centos7 and does not pertain to the default value but to the mechs of changing it that didn't work. We can change the value before merging to contain the default. I hadn't looked at that yet. It was just a guestimate.

it looks like there is no such file for debian/ubuntu. we can skip it for now (better to have).

currently the value for rhel is 4096 which is too small.

$ cat packaging/el8/cloud.limits
cloud hard nofile 4096
cloud soft nofile 4096

[weizhouapache]

code lgtm
not tested yet

the only concern is the value of open file descriptors

[blueorangutan]

Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 11761

[DaanHoogland]

@blueorangutan test keepEnv

[blueorangutan]

@DaanHoogland a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests

[blueorangutan]

[SF] Trillian test result (tid-11874)
Environment: kvm-ol8 (x2), Advanced Networking with Mgmt server ol8
Total time taken: 48279 seconds
Marvin logs: https://github.com/blueorangutan/acs-prs/releases/download/trillian/pr10040-t11874-kvm-ol8.zip
Smoke tests completed. 133 look OK, 0 have errors, 0 did not run
Only failed and skipped tests results shown below:

Test	Result	Time (s)	Test File

[DaanHoogland]

ping @vishesh92 , I think this is ready (if you agree)

[vishesh92]

We also need to set LimitNPROC for the management server to update max number of processes.

[vishesh92]

@blueorangutan package

[blueorangutan]

@vishesh92 a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress.

[blueorangutan]

Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 11840

[DaanHoogland]

@blueorangutan test

[blueorangutan]

@DaanHoogland a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests

[blueorangutan]

[SF] Trillian test result (tid-11939)
Environment: kvm-ol8 (x2), Advanced Networking with Mgmt server ol8
Total time taken: 47228 seconds
Marvin logs: https://github.com/blueorangutan/acs-prs/releases/download/trillian/pr10040-t11939-kvm-ol8.zip
Smoke tests completed. 133 look OK, 0 have errors, 0 did not run
Only failed and skipped tests results shown below:

Test	Result	Time (s)	Test File

[vishesh92]

clgtm. didn't test.

[vishesh92]

Tested. after the change

...
Max processes             unlimited            unlimited            processes
...

[weizhouapache]

is setting for LimitNOFILE needed ?
@DaanHoogland @vishesh92

[DaanHoogland]

I think none of the sessions are needed @weizhouapache , just the facility for operators to do it. The old method/file did not work with systemd, hence the new one.

[weizhouapache]

I think none of the sessions are needed @weizhouapache , just the facility for operators to do it. The old method/file did not work with systemd, hence the new one.

thanks @DaanHoogland

the file name filelimit.conf is a bit out-of-date.
never mind, let's keep it

[DaanHoogland]

the file name filelimit.conf is a bit out-of-date. never mind, let's keep it

ai, you are right. I'll create a rename PR soon.