[pve-devel] Feature Request: Enhancing Error Messaging

[pve-devel] Feature Request: Enhancing Error Messaging

[Rovshan Pashayev via pve-devel]

[-- Attachment #1: Type: message/rfc822, Size: 11595 bytes --]

From: Rovshan Pashayev <Rovshan.Pashayev@veeam.com>
To: "pve-devel@lists.proxmox.com" <pve-devel@lists.proxmox.com>
Subject: Feature Request: Enhancing Error Messaging
Date: Tue, 2 Jul 2024 10:50:09 +0000
Message-ID: <PH0PR14MB4735E67E7B94811CBBB3CCDB9FDC2@PH0PR14MB4735.namprd14.prod.outlook.com>

Dear Proxmox VE Team,

We kindly request the implementation of enhanced error messaging. Currently, all errors and failures are combined into a single string, making it difficult to identify and address specific issues.

Please consider introducing error codes or a similar mechanism to provide a structured format for error reporting. This would improve the user experience and expedite troubleshooting during third-party backup processes.

Thank you for your attention

Best regards, 
Rovshan Pashayev, Analyst, Veeam Agent for Linux, Mac, AIX & Solaris


[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] Feature Request: Enhancing Error Messaging

[Fabian Grünbichler]

> Rovshan Pashayev via pve-devel <pve-devel@lists.proxmox.com> hat am 02.07.2024 12:50 CEST geschrieben:

> We kindly request the implementation of enhanced error messaging. Currently, all errors and failures are combined into a single string, making it difficult to identify and address specific issues.

which errors? for which actions/API endpoints/..?

> Please consider introducing error codes or a similar mechanism to provide a structured format for error reporting. This would improve the user experience and expedite troubleshooting during third-party backup processes.

we try to keep the errors meaningful and easy to understand. if you have concrete examples that could benefit from more detail or changed contents, feel free to file issues at https://bugzilla.proxmox.com

we are not planning to add error codes (like E1234) to our error messages at the moment - in our experience, those don't provide much added benefit at the level of interaction PVE usually provides. 

structured error types can make sense for low level interfaces (where callers might decide on a course of action based on the error, like retrying). note that we do differentiate between lacking permissions, bad parameters and other errors also on the API level.


_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] Feature Request: Enhancing Error Messaging

[Rovshan Pashayev via pve-devel]

[-- Attachment #1: Type: message/rfc822, Size: 13329 bytes --]

From: Rovshan Pashayev <Rovshan.Pashayev@veeam.com>
To: "Fabian Grünbichler" <f.gruenbichler@proxmox.com>, "Proxmox VE development discussion" <pve-devel@lists.proxmox.com>
Subject: Re: [pve-devel] Feature Request: Enhancing Error Messaging
Date: Tue, 2 Jul 2024 13:02:23 +0000
Message-ID: <PH0PR14MB473525099909EC233147C0759FDC2@PH0PR14MB4735.namprd14.prod.outlook.com>

Hello Fabian,

Example:
If we attempt to start VM with problematic storages from WebUI, we might see following error (example):
https://i.ibb.co/Lx5QjCg/image-2.png

But if we attempt start action via SSH request we get both output in the one line (example):
WARN: no efidisk configured! Using temporary efivars disk. storage 'NFS_pve-storage' does not exist

Which causes confusion, during differentiating errors.

________________________________________
From: Fabian Grünbichler <f.gruenbichler@proxmox.com>
Sent: Tuesday, July 2, 2024 14:19
To: Proxmox VE development discussion <pve-devel@lists.proxmox.com>
Cc: Rovshan Pashayev <Rovshan.Pashayev@veeam.com>
Subject: Re: [pve-devel] Feature Request: Enhancing Error Messaging


> Rovshan Pashayev via pve-devel <pve-devel@lists.proxmox.com> hat am 02.07.2024 12:50 CEST geschrieben:

> We kindly request the implementation of enhanced error messaging. Currently, all errors and failures are combined into a single string, making it difficult to identify and address specific issues.

which errors? for which actions/API endpoints/..?

> Please consider introducing error codes or a similar mechanism to provide a structured format for error reporting. This would improve the user experience and expedite troubleshooting during third-party backup processes.

we try to keep the errors meaningful and easy to understand. if you have concrete examples that could benefit from more detail or changed contents, feel free to file issues at https://bugzilla.proxmox.com/

we are not planning to add error codes (like E1234) to our error messages at the moment - in our experience, those don't provide much added benefit at the level of interaction PVE usually provides.

structured error types can make sense for low level interfaces (where callers might decide on a course of action based on the error, like retrying). note that we do differentiate between lacking permissions, bad parameters and other errors also on the API level.


[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Re: [pve-devel] Feature Request: Enhancing Error Messaging

[Fabian Grünbichler]

> Rovshan Pashayev <rovshan.pashayev@veeam.com> hat am 02.07.2024 15:02 CEST geschrieben:
> Hello Fabian,
> 
> Example:
> If we attempt to start VM with problematic storages from WebUI, we might see following error (example):
> https://i.ibb.co/Lx5QjCg/image-2.png
> 
> But if we attempt start action via SSH request we get both output in the one line (example):
> WARN: no efidisk configured! Using temporary efivars disk. storage 'NFS_pve-storage' does not exist
> 
> Which causes confusion, during differentiating errors.

please post "pveversion -v" and the contents of the corresponding task file for that start task (/var/log/pve/tasks/...). it seems very likely that this issue is on your end (i.e., with the terminal/shell/.. you are using?)).

the screenshot would indicate that the task log is correct. if I manually try to provoke the same error and run "qm start XXX" in a shell, there is a proper new line between the warning line and the final line with the fatal error:

$ qm start 740001
generating cloud-init ISO
WARN: no efidisk configured! Using temporary efivars disk.
storage 'ext3' does not exist

note that everything but the last line is printed on STDOUT, while the last line is printed on STDERR..


_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel