Mysterious Stack Failure/s

Currently reading
Mysterious Stack Failure/s

248
15
Operating system
  1. macOS
Mobile operating system
  1. iOS
Last edited:
Hello everyone,

The media server has been up and running for the past two days as expected. Suddenly, with no known environmental change and on its own, every app inside the stack fails. Docker gives "Fail to run. No such container" type errors. Portainer gives "Created" status for 5 apps, "Existed" for 2....as if the Stack itself crashed/corrupted.

Logs
• No logs for the 5 Created status apps (jackett, transmission, prowlarr, jellyfin, lidarr)
• Logs for Sonarr and Radarr are below, nothing obvious that I'm seeing.
Adding as an image to zoom to spare everyone the wall of text.
Imagine there is a way to paste logs that not part of the primary view w/a tag?


1683721082067.png


1683721238639.png


1683721560683.png

Obviously the Created have today's timestamp. No idea why or how this happened...

1683721645997.png

Please ignore homebridge...
 
Aaaand...another docker-related wall of unfiltered spaghetti for some sucker to parse...Ughhh :confused:
 
Last edited:
Aaaand...another docker-related wall of unfiltered spaghetti for some sucker to parse...Ughhh :confused:
Sorry if it comes across that way as I'm trying to be thorough. Thought this was rather well broken down so, respectfully, asking for clarification around how this could be better filtered. Even provided screen shots so there wouldn't be some wall of text. I am not a code guy but doing what I can, putting in the effort. I did fix an issue from a log references a few days back.

While I expect rebuilding the stack will fix the issue, that fix will be temporary and looking for its root cause of the stack break.


Rebuilding the stack has it all running again...and expecting the stack to crash in the next 48-60 hours.
(Working 05/10/23 @ 11:30am)
 
Sorry if it comes across that way as I'm trying to be thorough. Thought this was rather well broken down so, respectfully, asking for clarification around how this could be better filtered. Even provided screen shots so there wouldn't be some wall of text. I
You added these after I posted. Originally, there were just 2 walls of undelineated text.


This is - at least - the third thread you've started this week on what amounts to the same docker issue, over & over. Numerous experienced members of the forum have devoted what must amount to hours of their time to your docker problems. You concluded the latest thread less than 24hrs ago, and now you've started yet another one up.


You did the same on threads around backups & encrypted drives, essentially coming back and asking the same question all over again despite people answering you and giving you perfectly good suggestions and fixes. Do you actually read all the help people post? It appears you don't.

Rather than presenting folks with a spaghetti of duplicated and failed docker containers, logs, weirdly formatted error messages and gui screengrabs (and then seemingly ignoring half of the suggestions as though they havent been made) why not start from basics and build up your containers one at a time. Be systematic. Get one working, test it, then move onto the next. That way, you have a chance of actually learning about how docker works, how mapped volumes work, how portainer interacts with them all, rather than poking a stick into random parts of the undifferentiated mess you've created in an attempt to chase down problems.
 
Last edited:
I see that you use Watchtower.
From my little experience, Watchtower gives troubles when updating multiple containers created by a single stack. I don't know why, never tried to figure out because I decided to update containers myself if needed.
This could be a reason for your issue.

If the containers do not depend on each other I believe it is a better practice to create a separate stack for each one.
On the other hand, if containers need to be in a single stack (let's say GlueTun with containers depending on it) then it is better to not let watchtower update them automatically.

This is just a comment, I am not sure that this is the cause of the problem, but it may help.
 
I see that you use Watchtower.
From my little experience, Watchtower gives troubles when updating multiple containers created by a single stack. I don't know why, never tried to figure out because I decided to update containers myself if needed.
This could be a reason for your issue.

If the containers do not depend on each other I believe it is a better practice to create a separate stack for each one.
On the other hand, if containers need to be in a single stack (let's say GlueTun with containers depending on it) then it is better to not let watchtower update them automatically.

This is just a comment, I am not sure that this is the cause of the problem, but it may help.
Appreciate that insight. If the stack crashes I'll experiment w/Watchtower. Thank you.
 
You added these after I posted. Originally, there were just 2 walls of undelineated text.

This is - at least - the third thread you've started this week on what amounts to the same docker issue, over & over. Numerous experienced members of the forum have devoted what must amount to hours of their time to your docker problems. You concluded the latest thread less than 24hrs ago, and now you've started yet another one up.

You did the same on threads around backups & encrypted drives, essentially coming back and asking the same question all over again despite people answering you and giving you perfectly good suggestions and fixes. Do you actually read all the help people post? It appears you don't.

Rather than presenting folks with a spaghetti of duplicated and failed docker containers, logs, weirdly formatted error messages and gui screengrabs (and then seemingly ignoring half of the suggestions as though they havent been made) why not start from basics and build up your containers one at a time. Be systematic. Get one working, test it, then move onto the next. That way, you have a chance of actually learning about how docker works, how mapped volumes work, how portainer interacts with them all, rather than poking a stick into random parts of the undifferentiated mess you've created in an attempt to chase down problems.
Does one respond to a tit-for-tat call out or not, as these generally run into diatribes and rarely will one convince the other of their perspective? Choosing yes, primarily because I respect this board a great deal and those who have been so helpful.I apologize if these posts frustrated you and hope, while you may not agree, you will have a more informed understanding from another perspective.

Spaghetti Wall
Think a good number of people reading will agree the posts were thoughtful, well formatted and written with consideration of the reader in mind; an easy read. One post had the logs pasted in, using a code tag which I expected would tidy up upon posting. That was not the case so I switched over the images for logs as makes for a tidier post. Interesting as I am now seeing for the first time the spoiler tag, something that might have worked on the earlier posts.
Note that I will refine posts for more clarity or information, partly because that is the encouraged behavior from the forum as posts are often auto merged...so timestamps aren't gong to give the true history of events cited by the post. I would even go back and place an official end post to assist in the topic closure and to say thank you.​
Undelineated not.

Topics
It appears you view all of these posts as the same, categorized as all Docker. Docker is a large space, and each of the posts pertained to a different aspect of Docker. Updating, Portainer problem, stack crashing.​
Backup & Encryption are Synology, not docker. I am not in this space day after day, ideally set and forget.​
If something goes wrong, it is often the case more than one thing goes wrong. Yep, I google read posts, experiment and will post for input.​
Effort
If one were to go back and review interactions I think one would be hard pressed to find a post to which I have not responded; even this one. One small post was not seen straight away and I responded to it as well. There were a few responses which seemed spot on and required a level of knowledge (low as it might be) to truly understand how to incorporate...so rather than risk exasperating the forum (which is what I'm seeing now) I needed to find other ways to fix which were understood different ways.​
When learning about docker and its cool application I spent time learning, even had homebridge up and running. it was a few years ago now, can't say I'd have it all memorized. Everything worked...saved it all in multiple ways to prevent the amount time the project consumed. While I understand some segments, and 'how' mapped volumes work - I know I'd have a hard time remapping if starting from scratch. Too much went into the media set up to start over. Returning to the media set up and seeing it broken, it didn't seem like a recreation from scratch would be necessary. Updates and tweaks. Turns out it was some work, was frustrating but understand that better now for next time...AND...thankfully did not need to recreate everything from scratch.​
The posts were in sequential order of the issues experienced. Docker services not working...discovered Portainer problem which became its own mini project (and post,) thought it was all good to go and then hit by the failing stack.​

I humbly apologize to anyone frustrated by these posts (primarily @Rusty who has been fantastic and clearly good person,) and truly appreciate the support those who have responded provided - even Fortran for possibly verbalizing what others have been thinking.

giphy.gif
 

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account on our community. It's easy!

Log in

Already have an account? Log in here.

Similar threads

Most likely a bit-flip. I would not want to complete multiple drive-writes on an SSD for fun. Probably I...
Replies
11
Views
1,388
For failure rate, the backblaze article on the topic is interesting...
Replies
6
Views
796
And this? Can anything other than your MBP connect to the problem wifi(s), eg your phone, another laptop?
Replies
15
Views
1,468
Deleted member 5784
D
Checking the below with the auto power on is what I use and it works. I think the “auto power on” is what...
Replies
3
Views
13,523

Welcome to SynoForum.com!

SynoForum.com is an unofficial Synology forum for NAS owners and enthusiasts.

Registration is free, easy and fast!

Trending threads

Back
Top