posts/snapraid-14.0-and-new-daemon/

[giscus[bot]]

posts/snapraid-14.0-and-new-daemon/

If you’ve been following this blog for a while, you know SnapRAID is central to my homelab storage setup. I’ve been running it for years on top of MergerFS to protect my media array, and I even maintain a SnapRAID sync script that a lot of you have been using. So when SnapRAID 14.1 dropped alongside the brand-new SnapRAID Daemon, I had to check it out.

https://zackreed.me/posts/snapraid-14.0-and-new-daemon/

[Harpz]

Hi

Great guide as always, I'm trying to set up email but its throwing an error.

Gone through the steps as above

sudo nano /root/.netrc

added:
machine smtp.gmail.com
login h***gmail.com
password googlesuppliedpasswordfromlink

Changed the permissions:
sudo chmod 600 /root/.netrc

Uncommented the notify_result line and changed to:
notify_result = curl --netrc --url smtps://smtp.gmail.com:465 --ssl-reqd --mail-from h***@gmail.com --mail-rcpt h***@gmail.com --upload-file -

Also changed: notify_result_level = info

Restarted service:
sudo systemctl restart snapraidd

Error:

Exit Code: -1
syslog: command curl --netrc --url smtps://smtp.gmail.com:465 --ssl-reqd --mail-from h****@gmail.com --mail-rcpt h*****@gmail.com --upload-file - terminated in 2 seconds with exit code 55
syslog: failed to send error report
Notification failed (check syslog for details)

Any pointers?

[rubylaser]

Hi! I'm really glad that this post was helpful 😄 Let's try to get you working.

An exit code 55 in curl mean that it's unable to send network data. This is almost always going to be something to do with the contents of your .netrc file. It almost always means curl couldn't read the .netrc file or there's some formatting issue with it .

First, let's make sure that you are the root user and that the file is in the right spot (it looks like it is based on what you posted, but let's double check first).

1. Verify the file is actually where curl will look for it:

sudo -i
cat /root/.netrc

Hopefully, this returns the contents you posted above

2. Check .netrc file ownership. It needs to be owned by root, not the sudo user:

ls -la /root/.netrc

Should show -rw------- 1 root root. If it shows you username as owner, fix it:

chown root:root /root/.netrc

3. Check the .netrc format carefully the app password Google provides has spaces in it (formatted as 4 groups of 4 characters). The .netrc file needs the password without spaces:

machine smtp.gmail.com
login h****@gmail.com
password abcdefghijklmnop

Not:

password abcd efgh ijkl mnop

4. Verify curl can find it manually:

curl --netrc --url smtps://smtp.gmail.com:465 --ssl-reqd  --mail-from h****@gmail.com --mail-rcpt h****@gmail.com --upload-file - <<< "Subject: test"

5. Running the command as the correct user?:
Finally, make sure that you are running the command as the root user in the snapraid deamon, not as nobody.

I'm betting it's the spaces in your Gmail app password. Try these things out and let me know if it's solved.

[Harpz]

Hi,

All set as it should be, I had even taken the spaces out of the supplied google password.

Its working now, i had:

notify_run_as_user = nobody

Soon as I changed it to:

notify_run_as_user = root 

All appears to work, thank you so much for helping so quickly, may be worth popping a note in the guide so doughnuts like me dont miss silly things like this :-)

[rubylaser]

Awesome! I'm glad that it's working now!

Thanks for the idea; I've added a note to run it as root, so hopefully that doesn't catch others in the future.

I've also added a new section this morning that covers pausing/unpausing Docker containers if you use them.

[Harpz]

I don't use Docker at the moment but been thinking about looking into it, its great to see the guide growing as things get added / discovered.

Random question as you've been playing with this longer then I, if i choose to clear some files out on the array and it exceeds the max allowed deleted files, is there a way in the web interface to force a sync, cant seem to find it.

As it stands looks like my old SnapRAID menu script still has a use :) will just need to modify it to stop the daemon when I'm using it so they don't clash

[rubylaser]

Andrea mentioned that later versions of the web interface will have additional options to manually run jobs, but as it stands right now, you can't temporarily exceed the delete thresholds for a one time job. Although, you could temporarily bump up the number in the web interface for a night.

Honestly, I'd just run a snapraid sync from the command line one-time by one-off jobs like this.

[brkirkpatrick]

Been using hook script based on your example, and all of a sudden it is failing with the following error:
The pre_run_script terminated with exit code 1
Log File: /var/log/snapraid/20260629-134902-sync.log
syslog: script [path]/snapraid-hook.sh terminated in 10 seconds with exit code 1

Script as follows:
#!/bin/bash

case "$1" in
task-begin)
docker pause sonarr radarr nzbget
;;
task-end|task-error)
docker unpause sonarr radarr nzbget
;;
esac

For some reason it is failing to pause radarr, both sonarr and nzbget pause appropriately. What is also strange is that while radarr container is not showing as paused, something is happening as the webui stops responding after the error is thrown. Containers pause appropriately if I run the command manually. I've tried removing and recreating all of the containers multiple times, in addition to using different versions/tags.

Any thoughts/suggestions?

[rubylaser]

I'm wondering if the first container is taking slightly too long to pause and is fast failing. Can you introduce a little delay to see if that works?

#!/bin/bash

CONTAINERS="sonarr radarr nzbget"

case "$1" in
  task-begin)
    for c in $CONTAINERS; do
      docker pause "$c"
      sleep 5
    done
    ;;
  task-end|task-error)
    for c in $CONTAINERS; do
      docker unpause "$c" 2>/dev/null
    done
    ;;
esac

[brkirkpatrick]

I had a similar thought, though wasn't sure the best way to add the delay, will try your suggestion.

That said, it is the middle container in the list that was failing to pause :/. Will try in any case and report back.

[brkirkpatrick]

Since I swapped the order of the containers in the script, things had been working. Coincidentally, the script failed again last night, so I just added the delays per your suggestion and on first run it worked! Will keep monitoring, seems a strange failure but fingers crossed the delay works.

[brkirkpatrick]

Working fine for the last week with delay as suggested. Thanks for the assist!

[rubylaser]

Working fine for the last week with delay as suggested. Thanks for the assist!

I’m glad that fixed it! Thanks for letting me know!