Forwarded to devs Advanced monitoring stopped working after Grafana 7.0.0 update. Ubuntu/Centos

[IronDonDon]

User name: IronDonDon

TITLE

Advanced monitoring stopped working after Grafana 7.0.0 update. Ubuntu/Centos

PRODUCT, VERSION, OPERATING SYSTEM, ARCHITECTURE

Ubuntu 18.04.4, Centos 7.8.2003

PROBLEM DESCRIPTION

Plesk updated system packages, which included grafana server 7.0.0, after which Advanced monitoring stopped working. Downgrading grafana server to version 6.7.3 solves the issue.

STEPS TO REPRODUCE

Update/upgrade grafana to 7.0.0

ACTUAL RESULT

Advanced monitoring not working.

EXPECTED RESULT

Monitoring should work. No error must be produced.

ANY ADDITIONAL INFORMATION

t=2020-05-19T02:16:04+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=3 name="Mail server memory usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:09+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=4 name="MySQL memory usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:12+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=6 name="Partition \"/\" utilization" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:15+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=5 name="Plesk memory usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:16+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=8 name="Swap usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:20+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=7 name="Real memory usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:31+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=1 name="Apache & PHP-FPM memory usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:31+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=4 name="MySQL memory usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:32+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=3 name="Mail server memory usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state
t=2020-05-19T02:16:33+0200 lvl=eror msg="Alert Rule Result Error" logger=alerting.evalContext ruleId=2 name="nginx memory usage" error="tsdb.HandleRequest() error Could not find executor for data source type: grafana-simple-json-backend-datasource" changing state to=keep_state

YOUR EXPECTATIONS FROM PLESK SERVICE TEAM

Confirm bug

 

[IgorG]

Thank you for report. Bug EXTPLESK-1867 has been already created.

As a possible workaround you can add a new section to the grafana configuration file /etc/grafana/grafana.ini

[plugins]

allow_loading_unsigned_plugins = grafana-simple-json-backend-datasource

and restart grafana server

# service grafana-server restart

On Windows the configuration file is here: C:\Program Files (x86)\Plesk\var\modules\grafana\grafana\conf\custom.ini
Grafana server on Windows/Unix also can be restarted via the command:

plesk ext grafana --grafana-restart

 

[UHolthausen]

Hello

since the update of grafana yesterday the graphic remains empty.

OS debian 9.12
plesk obsidian 18.0.27

 

[MicheleB]

Same behaviour on CloudLinux 7.8 and Plesk 18.0.27.
Fortunately I see that the issue is also present on other operating systems (CloudLinux receives from Plesk non-priority support over Ubuntu/CentOS) so I assume that there's some chance to see it fixed soon, at least I hope.

 

[akira9000]

Just seen the issue this morning but before I'd seen this thread. Had tried logging out of Grafana but now unable to log back in - just get the login.OAuthLogin(missing saved state) alert and Advanced monitoring shows the attached error. Can anybody tell me how to at least get back into Grafana? Advice appreciated.

I'm on Centos 7.8.2003 & Obsidian 18.0.27

 

[MicheleB]

The issue was fixed with the last update released a few minutes ago (at least on my Plesk installation).

 

[IgorG]

The issue was fixed with the last update released a few minutes ago (at least on my Plesk installation).

Yes, Plesk extension Grafana was updated with fix Grafana - Plesk Extensions

 

[UHolthausen]

Hello

I cannot confirm. Still the same bug
OS: Debian 9.12
possibly an uninstallation and following reinstallation helps ---> no, didnt work

the same issue as before

 

[pould]

Thanks Igor - I can't see an update to the Extension yet but I assume it will be available in the next hour or so. I'll check back then and update.

 

[pould]

Update from me - I can see I have the most recent version of Grafana installed (1.1.5-163) but still have the bug here

CentOS Linux 7.8.2003

 

[Werner99]

For me it works, updated the extension to 1.1.5-163 and now everything is back to normal on Debian 9.12

 

[hrvoje]

Plesk updated today to Version 18.0.27
Grafana Extension to 1.1.5-163
CentOS Linux 7.8.2003 (Core)


Did grafana server restart (service grafana-server restart), still having this problem

 

[AYamshanov]

An error message like "login.OAuthLogin (missing saving state)" (https://talk.plesk.com/attachments/screenshot-2020-05-19-at-10-09-24-png.16924/) could be shown if a page generating process took more than 60 secs (it is a timeout for an auth token), usually, it could be fixed by page refreshing.

Could you please check "/var/log/grafana/grafana.log" for any error messages?

 

[UHolthausen]

Hello

could be fixed by page refreshing.

sorry, but a refresh or a reload of the page always brings the same error
Extract grafana log see appendix

 

[AYamshanov]

t=2020-05-19T19:19:07+0200 lvl=eror msg="Failed to look up user based on
cookie" logger=context error="user token not found"
t=2020-05-19T19:19:07+0200 lvl=eror msg="Failed to look up user based on
cookie" logger=context error="user token not found"

Hm... this error could be caused if an old cookie with session id is used. Could you please check how Advanced Monitoring/Grafana works in an incognito mode in a browser? If it works, I suggest deleting browser cookies; to be more precision, you need to delete "grafana_session" cookie:

 

[UHolthausen]

Hello

that was it !! Its working now.




edit:

unfortunately only helped for a short time. Just logged in again and same error as before.
Deleting the cookie grafana session helps, but if you have to do that every time, it's annoying.

 

[akira9000]

Clearing cookies fixed mine.

 

[hrvoje]

clearing cookies fixed problem here too...

 

[AYamshanov]

It is good to know that it resolved most of the issues. Thank you for the feedback.

@UHolthausen, I suggest contacting our Plesk Support team because it is a little bit strange and maybe we need to investigate the issue on your server environment.

 

[UHolthausen]

Hello

thanks for the help.
Again deleted the cookies of the Grafana session and restarted the server completely.
So far it works and the graphic is displayed.
Sometimes the devil is in the details.