[GH-ISSUE #3072] High Cpu Usage after new installation during this week #1753

Closed
opened 2026-02-27 11:18:56 +03:00 by kerem · 8 comments
Owner

Originally created by @samuraikid0 on GitHub (Sep 22, 2023).
Original GitHub issue: https://github.com/modoboa/modoboa/issues/3072

Impacted versions

  • OS Type: Debian
  • OS Version: Bullseye
  • Database Type: PostgreSQL
  • Database version: 13
  • Modoboa: Last
  • installer used: Yes
  • Webserver: Nginx

Steps to reproduce

Complete a fresh installation and look at task manager you see amavis foreground process run 99% all the time.

Current behavior

Expected behavior

System don't run at 100%

Video/Screenshot link (optional)

Originally created by @samuraikid0 on GitHub (Sep 22, 2023). Original GitHub issue: https://github.com/modoboa/modoboa/issues/3072 # Impacted versions * OS Type: Debian * OS Version: Bullseye * Database Type: PostgreSQL * Database version: 13 * Modoboa: Last * installer used: Yes * Webserver: Nginx # Steps to reproduce Complete a fresh installation and look at task manager you see amavis foreground process run 99% all the time. # Current behavior <!-- This occurred during this week, last week install I've had a normal system --> # Expected behavior System don't run at 100% # Video/Screenshot link (optional)
kerem 2026-02-27 11:18:56 +03:00
  • closed this issue
  • added the
    stale
    label
Author
Owner

@samuraikid0 commented on GitHub (Sep 22, 2023):

high cpu

What I've notice during installation

gevents takes 6x more time to build

##############################

Creating temporary files...

Recursively querying SPF records of selected whitelist mailers...

Querying webmail hosts...
Getting spf.messagingengine.com

Querying social network hosts...

Querying ecommerce hosts...

Querying bulk mail hosts...

Querying miscellaneous hosts...

Querying custom hosts...

Including scraped Yahoo! outbound hosts...

Removing invalid IPv4 CIDRs from whitelist...................................................................................................................

Sorting whitelist rules...

Writing 1063 whitelist rules to /etc/postfix/postscreen_spf_whitelist.cidr...

Reloading Postfix configuration to refresh rules...

Done!

After install the dkim keeps red with not found all other are green I'll give it more time to see if it changes

<!-- gh-comment-id:1732085569 --> @samuraikid0 commented on GitHub (Sep 22, 2023): ![high cpu](https://github.com/modoboa/modoboa/assets/46766062/d6bed86d-d95b-4f91-b574-d08a660d0b56) What I've notice during installation `gevents takes 6x more time to build` ############################## ``` Creating temporary files... Recursively querying SPF records of selected whitelist mailers... Querying webmail hosts... Getting spf.messagingengine.com Querying social network hosts... Querying ecommerce hosts... Querying bulk mail hosts... Querying miscellaneous hosts... Querying custom hosts... Including scraped Yahoo! outbound hosts... Removing invalid IPv4 CIDRs from whitelist................................................................................................................... Sorting whitelist rules... Writing 1063 whitelist rules to /etc/postfix/postscreen_spf_whitelist.cidr... Reloading Postfix configuration to refresh rules... Done! ``` After install the dkim keeps red with not found all other are green I'll give it more time to see if it changes
Author
Owner

@tonioo commented on GitHub (Sep 28, 2023):

@samuraikid0 Have you tried to restart this process?

<!-- gh-comment-id:1739072565 --> @tonioo commented on GitHub (Sep 28, 2023): @samuraikid0 Have you tried to restart this process?
Author
Owner

@samuraikid0 commented on GitHub (Sep 28, 2023):

@samuraikid0 Have you tried to restart this process?

Yes, I've even purge amavisd and dkim and run the upgrade again attempting to reset the process high cpu without luck.

I'll wait to see if someone else reports thee same issue

Thanks !

<!-- gh-comment-id:1739409618 --> @samuraikid0 commented on GitHub (Sep 28, 2023): > @samuraikid0 Have you tried to restart this process? Yes, I've even purge amavisd and dkim and run the upgrade again attempting to reset the process high cpu without luck. I'll wait to see if someone else reports thee same issue Thanks !
Author
Owner

@stale[bot] commented on GitHub (Mar 17, 2024):

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

<!-- gh-comment-id:2002461616 --> @stale[bot] commented on GitHub (Mar 17, 2024): This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
Author
Owner

@Relic180 commented on GitHub (Oct 7, 2024):

I'm experience this as well, with a fairly similar build to OP:

  • OS Type: Ubuntu
  • OS Version: 24.04 (Noble Numbat)
  • Database Type: PostgreSQL
  • Database version: 17
  • Modoboa: Last (Oct 7th)
  • installer used: Yes
  • Webserver: nginx (1.24)

Immediately after the install completed, my CPU usage shows up consistently around 80%. Prior to install, it had been between 5-15%.

I installed it yesterday originally. Overnight my VPS host capped me and sent a bunch of notifications / warnings for running at full-tilt for over 12 hours straight. Everything got stuck and unresponsive, I had to get uncapped just so I could restore from a backup image.

Trying to install again, and looks like I'm seeing the exact same thing. Restarted a couple times to be sure. CPU always hovering between 75-85%. Most terminal commands take far longer to execute than normal, etc...

<!-- gh-comment-id:2398095253 --> @Relic180 commented on GitHub (Oct 7, 2024): I'm experience this as well, with a fairly similar build to OP: - OS Type: Ubuntu - OS Version: 24.04 (Noble Numbat) - Database Type: PostgreSQL - Database version: 17 - Modoboa: Last (Oct 7th) - installer used: Yes - Webserver: nginx (1.24) Immediately after the install completed, my CPU usage shows up consistently around 80%. Prior to install, it had been between 5-15%. I installed it yesterday originally. Overnight my VPS host capped me and sent a bunch of notifications / warnings for running at full-tilt for over 12 hours straight. Everything got stuck and unresponsive, I had to get uncapped just so I could restore from a backup image. Trying to install again, and looks like I'm seeing the exact same thing. Restarted a couple times to be sure. CPU always hovering between 75-85%. Most terminal commands take far longer to execute than normal, etc...
Author
Owner

@tonioo commented on GitHub (Oct 8, 2024):

Can you check which process is consuming your CPU(s) ?

<!-- gh-comment-id:2399441492 --> @tonioo commented on GitHub (Oct 8, 2024): Can you check which process is consuming your CPU(s) ?
Author
Owner

@cyberarm commented on GitHub (Oct 8, 2024):

Have you checked that you're not using port 9000?
IIRC, I had a high cpu issue due to using that port and the modoboa worker attempting to bind to it, failing, and getting stuck in a loop.

<!-- gh-comment-id:2399993663 --> @cyberarm commented on GitHub (Oct 8, 2024): Have you checked that you're not using port 9000? IIRC, I had a high cpu issue due to using that port and the modoboa worker attempting to bind to it, failing, and getting stuck in a loop.
Author
Owner

@Relic180 commented on GitHub (Oct 8, 2024):

Although what ultimately kicked-off the behavior was the Modoboa installation, the issue was actually with pm2.

Not entirely sure what happened, but after the install pm2 went and fetched an outdated ecosystem config that had some no-longer-valid paths in it. So as soon as the install finished, pm2 started throwing massive errors and devouring my CPU.

Once I found it and fixed the paths, everything went back to normal. Sorry about the false alarm.

<!-- gh-comment-id:2400049890 --> @Relic180 commented on GitHub (Oct 8, 2024): Although what ultimately kicked-off the behavior was the Modoboa installation, the issue was actually with pm2. Not entirely sure what happened, but after the install pm2 went and fetched an outdated ecosystem config that had some no-longer-valid paths in it. So as soon as the install finished, pm2 started throwing massive errors and devouring my CPU. Once I found it and fixed the paths, everything went back to normal. Sorry about the false alarm.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
starred/modoboa-modoboa#1753
No description provided.