Have you had an issue with a server or a service recently? Or maybe you’re having one right now? If so, the quickest way to identify the root cause can be analyzing the log files.
Activating our server-wide Log Browser extension can make the life of a Plesk admin easier when it comes to use cases like:
- Investigating service logs and statuses
- Searching for errors and warnings in journald logs
- Filtering out for content of the Plesk panel.log file
- Investigating email-related issues.
Let me walk you through troubleshooting each one of these.
Investigate Service Logs and Statuses
I deliberately broke the config of one of the services on my server to show you how quickly you can identify the issue and figure out what went wrong.
From the item with the summary about the service, you can quickly jump to the service’s logs by clicking on the “list” icon on the right side of the line. Let’s also filter out only “error” messages from the log. As you can see in the image below, the issue is “/etc/named.conf:4: unknown option ‘option’“, which is true because I made a typo, writing “option” instead of “options”.
Searching for Errors and Warnings in Journald Logs
In the previous section, we investigated what went wrong with one particular service. But you can also check all system’s journald logs from the “System” tab. For example, I have found that my server was unavailable on 28 Nov 2022 between 14:26 and 14:53.
I’m interested in finding any log records for that period of time. I’ll try to find out what happened exactly at that time to see what system has reported the log files.
What do you see in the image below?
- 2022-11-28 14:26:56 – The system was finished with system power off operation. Sounds like somebody correctly switched the server off because everything was prepared for the system shutdown.
- 2022-11-28 14:52:26 – The system starts booting up with loading of the kernel.
This is true: I had switched off the server to test 360 Monitoring for another blog post and then powered it back on once the test was completed.
Investigating Email-related Issues
An email that delivered successfully
Here’s an example of an email that was delivered without issues from one email address to another. We can see from the text “dsn=2.0.0, status=sent (delivered via plesk_virtual service)” in the second line in the image below. After that, the email was removed from the mail service queue, as you can see in the first line below: (“9B85740A44: removed”).
The SMTP 25/tcp port is blocked
This is a really common case where a cloud/hosting provider blocks SMTP 25/tcp port by default to prevent sending spam. You try to send an email but it’s not delivered because the server can’t connect to external mail systems. Here’s an example of what it looks like in the server log files (“connect to […]:25: Connection timed out”).
Search an email by id “Unknown user”
If you know the id of an email inside the mailing service, you can track all its history, including adding to queue, checking, previous warning and current status. Here’s an example of an email that was sent to the wrong address (“Unknown user”) with the email bounceback that email could’’t be delivered.
Filter content within Plesk panel.log file
A developer case
If you are developing an extension or a script for Plesk, it can be useful to check logs for any warnings or errors to fix before the release. As you can see in the example below, it’s easy to spot PHP Warnings generated by the code.
When in doubt…
Whenever you don’t understand what’s broken and how to fix it, I recommend starting from the log files. Most software leaves clues or even detailed explanations of what went wrong. Sometimes, you can find an error message but it’s not clear what caused it. In such cases, I recommend turning to Plesk KB articles, Plesk Forum or search engines, since it’s likely that someone else has already faced the same issue.