Bug 2166213
Summary: | Unexpected CPU usage of sendmail | ||
---|---|---|---|
Product: | [Fedora] Fedora EPEL | Reporter: | mx <berlinli222> |
Component: | esmtp | Assignee: | Jaroslav Škarvada <jskarvad> |
Status: | NEW --- | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
Severity: | unspecified | Docs Contact: | |
Priority: | unspecified | ||
Version: | epel9 | CC: | jskarvad, olysonek-foss |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | x86_64 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | If docs needed, set a value | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | Type: | Bug | |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
mx
2023-02-01 07:32:31 UTC
Make sure you run esmtp and not different MTA: # alternatives --display mta Try removing the queue, it could be corrupted, the queue should be '$HOME/.esmtp_queue', $HOME should be /root in this case. it is estmp: [root@host .esmtp_queue]# alternatives --display mta mta - status is auto. link currently points to /usr/bin/esmtp-wrapper /usr/bin/esmtp-wrapper - priority 30 slave mta-sendmail: /usr/bin/esmtp-wrapper slave mta-sendmailman: /usr/share/man/man1/esmtp.1.gz slave mta-mailq: /usr/bin/esmtp-wrapper slave mta-mailqman: /usr/share/man/man1/esmtp.1.gz Current `best' version is /usr/bin/esmtp-wrapper. There is more than 20000 queue in 'root/.esmtp_queue'. It is no longer high cpu usage after rm all the queues in this folder. However, why all those queues are not send out? And there is empty in /var/spool/mail for root. note that, all those kind of info in this queue will send to /var/spool/mail/root in rhel7/centos7. I only installed estmp withput any manual configuration. I only installed estmp without any manual configuration. May we just send all those queues into /var/spool/mail/root by default ? Or just disable auto_send, do not send any thing in default. The queues in my machine are from the outputs of the jobs scripts enabled by crontab. Before solved of this issue, I can reduce the generation of those queues by dropping those outputs to /dev/null in `crontab -e`. Ok, default sending to local file, /var/spool/mail/root, is resolved by install esmtp-local-delivery: ``` dnf install esmtp-local-delivery ``` For users who want to delivery mails to their email, they should edit the configuration of esmtp. However I donot find any configuration files untill now, so can not mark any suggestion here. Are you sending the mail to local users or to remote users? Esmtp doesn't have queues, the queue support is script hack, but it should work. Because if you are sending the e-mail to local users you of course need some e-mail server listening, because esmtp is send only (hence its name): $ dnf repoquery --qf %{DESCRIPTION} esmtp ESMTP is a user configurable relay-only Mail Transfer Agent (MTA) with a sendmail-compatible syntax. It's based on libESMTP supporting the AUTH (including the CRAM-MD5 and NTLM SASL mechanisms) and the StartTLS SMTP extensions. You can either install some MDA and use the esmtp-local-delivery or install full e-mail server. So the e-mails are probably accumulating because esmtp is unable to connect to the e-mail server and the e-mails are remaining in the queue as undelivered. |