I will never understand why you have a problem with me. Nextcloud does not work and I come onhere to get assistance and every one just hates on me
Well, you don’t make it easy for us either. We are no clairvoyants. I and others have given you enough hints about what you can do to get help, but you chose to ignore them all. I can understand that you may not be able to answer everything. But you could at least describe in a few sentences how your setup looks like and what exactly you are trying to do. Without that we can’t help you. And yes it does come across a little arrogant when you just post four words in a user forum, and then expect people to help you. I’m sure that if you ask a friend in “real life” to help with something, you would at least add a please to it 
Let’s put together what‘s scattered across the forum:
NC / Server Info:
Nextcloud version (eg, 20.0.5) :
Nextcloud 21.0.7
Operating system and version (eg, Ubuntu 20.04) :Linux 5.4.0-92-generic x86_64
Apache or nginx version (eg, Apache 2.4.25) :replace me
PHP version (eg, 7.4) :7.4.3
(Server locked error attempt 2)
Problem description:
Then i was doing backups using duplicatai using the webdav and after a month the backups kept saying server locked. This was a next cloud issue and i made a second location and set that as backup destination and its been working. Any idea what or how it was locked preventing the job to work?
And
I am trying to do a backup to a directory on nextcloud using the webdav protocol. This worked for a bit until I kept see backup failed, server locked. I asked duplicati and they said def a next cloud. So to troubleshoot I made another directory and another backup using webdav like originally and thus so far the backups have been working. Any idea what happened with the original locale being locked? Or even what that means?
Ryzen 7 with radeon 580 and 16GB with a 1TB m.2
Proxmox
NGINX proxy
godaddy domain
letsencrypt
(Is it okay if i ask for help?)
I do not understand how it is okay for a single user not to be able to use nextcloud at all. Been trying to get help for months and it seems like that will not be happening. I am using duplicati to backup. Duplicati keeps returning a locked 423 error. No one has any idea why. Duplicati blames nextcloud and nextcloud blames my lack of information.
I remove the backup job and recreate it and it works fine. Then after a undisclosed amount of time it locks up again and I can not backup. I have no idea here.
And yes when it is locked I can use nextcloud all day without issue
I use duplicati to run backups. Duplicati keeps after a certain amount of backups says server is locked, 423 code. Acording to duplicati this is nextcloud issue. I have tried for months to get assistance and I can not. So now I am willing to pay someone to fix this.
(Willing to hire someone to fix server locked error)
Hope I haven’t missed anything important there …
So what I understand, that Nextcloud is indeed usable for you, but you have problems using Nextcloud as a Target for Duplicati backups, because after a while you receive WebDAV 423 (Locked) error. I don’t have any experience with using NC as target for Duplicati, so only suggestions. You might want to check what you use for file locking for Nextcloud (search for redis). Also check if you have any sync client active which might lock the files (Nextcloud desktop?!)
3 Likes
@SysKeeper nice work!
I’m not so sure that duplicati isn’t the troublemaker here… afaik it still is beta. at least duplicati v2.x - and hence backup up stuff with this tool is beta as well.
Maybe in general, but…
Hello. I am the person trying to support original poster on the Duplicati forum. You can search for @Adramelramalech there without login to find posting history. I did the same here before feeling a need to post anything. Meanwhile I spent time trying to learn Nextcloud better than I did before it became clear that I couldn’t explain why Nextcloud (or is that actually sabre/dav?) replied like this:
System.Net.WebException: The remote server returned an error: (423) Locked.
at Duplicati.Library.Main.BackendManager.Delete
In contrast to no other Duplicati report, a Google search for “Nextcloud” “423 locked” soon led to:
File is locked - how to unlock and this seemed a good generic answer to try or take to the experts.
The initial solution seemed to be to clear locks from database. If it kept happening, move to Redis.
This issue keeps happening (see above). Would anyone like to guide user through the exact steps?
I have not used Nextcloud, am fairly good at Duplicati, but am more on the forum than on GitHub which is where the developers (wish there were more) fight coding issues. Maybe you do this too?
The backends encapsulate the actual communication with a remote host with a simple abstraction, namely that the backend can perform 4 operations: GET, PUT, LIST, DELETE. All operations performed by Duplicati relies on only these operations, enabling almost any storage to be implemented as a destination.
What this means is that nothing fancy such as any WEBDAV file locking is tried, as far as I know.
Transactional file locking seems like something different. and manual says default uses database.
https://github.com/duplicati/duplicati/blob/master/Duplicati/Library/Backend/WEBDAV/WEBDAV.cs
if anyone wants a look. I’m not a C# developer but have done enough other coding to get an idea.
I can speak in general terms, but getting details on this specific problem has been hard (like here).
1 Like
How do I retrieve this required information?
Can you give more context?
Can we please just delete my topics?
Hey @ts678
Thank you very much for the insights! Just to be sure, do you know if redis is configured and used or was this just a suggested solution?
@Adramelramalech i am really glad that you have learned how to open new issues.
but it would be nice if you’d keep to at least one and try solving it first instead of opening a new one. If you haven’t noticed yet ppl here are REALLY trying to help you… but if you’s always start anew maybe other ppl would chime in and they would need all of your infos again (btw: @SysKeeper did a great job collecting at least SOME details about your server. that alone is an outstanding action and far from “why don’t you like me here on the forum?”)
you could install app issue template to your server, run ist and copy&paste the output here
this would give us many many informations…
but as @ts678 said… the problem could be somewhere else. As much as I understood from it it could only be solved by someone being able to code.
1 Like
Hey Jimmy look at our PM’s. Seriously?
I have no information on configuration. Someone could maybe lead user to check.
There seems to be a general problem with getting started, as per title of this topic.
The manual I cited says how to set config.php to Redis but needs prior knowledge.
At user forum level (if that is what this is), sometimes workaround is all one can get.
Basically, if one path (database locking) doesn’t survive, maybe use a different path.
Info about file locking is another post (from 2016) where using Redis helped locking.
It followed the path of the post I cited, starting with a clean up the file locks table fix.
This, though, assumes that someone knows how to get into the database to do that.
Server locked 423 error #30651
looks like OP effort on engaging Nextcloud server GitHub. It got closed, linking here.
You should be thankful to @JimmyKater and others here that they even trying to help you. It is not the task of forum participants to gather information from multiple threads and PMs to give you a concised answer. It is your task to describe your problem as concised as possible. Your problem is not a lack of technical skills, but a lack of communication skills combined with your demanding tone.
2 Likes
That is indeed a problem we have here as well. There were/are different attempts/suggestions to retrieve basic information about the problem and the environment.
To be fair, the issue you linked (Server locked 423 error · Issue #30651 · nextcloud/server · GitHub) suffers from the same problem, as does a issue posted by the same user on the duplicati issue tracker (https://github.com/duplicati/duplicati/issues/4614).
It’s still not clear to me (us?) how comfortable @Adramelramalech is with his server, the cli, etc. Therefore it’s hard to judge if setting up redis is easy or not. There’re different tutorials out there on how to install nextcloud and set-up redis, those might be a good starting point.
Checking if redis is configured can be checked in the config file (config/config.php). There should be a entry for memcache.locking. If accessing config/config.php is not suitable, the app issue template can be installed (as suggested by @JimmyKater), the config file with removed sensitive values can be retrieved there. Third option that comes to my mind would be the monitoring-api. This can be retrieved by the url /ocs/v2.php/apps/serverinfo/api/v1/info. In the case of @Adramelramalech this should translate to https://nc.zetta7networks.com/index.php/login?redirect_url=/ocs/v2.php/apps/serverinfo/api/v1/info I guess.
The Exception in duplicati comes from this part of the code (juding from the exception in the duplicati forum): https://github.com/duplicati/duplicati/blob/82c14c5cb77166393f73e0cfcb77e075faecb8c3/Duplicati/Library/Backend/WEBDAV/WEBDAV.cs#L283-L285 So duplicati will not treat the locked response any differently to other errors.
Sadly I’m not sure how to trigger the error, in a quick test doing a backup with duplicati to nextcloud all operations were successful. Of course this was just a test case and not running over multiple days.
A stumbled upon a few options in duplicati which might be worth a try, but just wild guesses…
no-connection-reuse, so each operation does use it’s own connection
number-of-retries, might be worth increasing? Not sure if the exception thrown prevents a retry though?
retry-delay, might also be worth increasing
synchronous-upload, might be worth a try.
Maybe @ts678 has an opinion about these options?
Thanks for trying so hard. I know there are other successful Nextcloud users (unknown count).
Retries can’t usually retry through a “solid” error, which this seems to be, based on long history.
What would have been nice would have been About → Show log → Live → Retry to watch error.
I made offers to help debug – didn’t get anywhere. Possibly more and longer retries could help.
synchronous-upload isn’t much used. I’m not sure how well it works, but it’s another wild guess.
A possible debug path would be first get a good look at the overall activity from the failure, then (assuming it’s always a delete 423) try other tools like Duplicati.CommandLine.BackendTool.exe.
Target URL can be copied from Export As Command-line. If it works, why did long sequence fail?
If it fails, then maybe next step would be to bypass WebDAV, and see if Nextcloud UI can delete.
None of this can be done without OP help. On Duplicati forum, OP here seems ready to give up.
If not, follow up there. Unfortunately that would leave an unsolved mystery, but those happen…
EDIT:
Actually, I wonder if Nextcloud UI would show a symbol indicating the file is locked? W2G2 does.
How to use W2G2 app in Nextcloud, but I don’t know if this is the same locking as WebDAV 423.
Regardless, one might want to get the file name from the Duplicati log to make UI search easier.
This is what I feared. So one way to improve this on the duplicati side would be to treat 423 as a soft error, instead of a hard error. But I don’t have enough knowledge of how things work in duplicati to judge if it is this easy.
Let’s assume for a second that there’s no bug in nextcloud and no bug in duplicati. Question is then if there’s anything which could lock the files. Any other sync clients or something. But this can only be answered by @Adramelramalech. One test could be to use another nextcloud account, just for backups. And see how this plays out.
It’s a soft error in the sense that things like short network outages and server-side glitches can be bypassed by retrying enough. If error recovery can’t get past them, the backup (or whatever) fails.
While it’s retrying, one doesn’t even hear about it. Maybe that’s a feature but it slows debugging…
Is there a rationale that a 423 be special-cased? The hard error isn’t the 423, it’s exhausting retries.
Yes. Problems frequently need help from their reporter in order to get anywhere. We’re all asking.
We’ll see if anything happens. If more clarification is required, specific polite questions might help.
Sorry, that’s how I understood your post ;-). If the normal retry works when a 423, I don’t see the need of a special handling. I understood retries were not done for a 423.
@Adramelramalech Thanks for the video! Can you please check, that 21.0.7 is indeed the running version? Check under Settings → Overview. I’m wondering, because that Oracle error should be fixed in 21.0.5 already (21.0.5 by blizzz · Pull Request #28987 · nextcloud/server · GitHub).