After the downtime in the last few days, most of my work was able to upload when things came back up.

However I have 6 work units that refuse to upload due to this Transient HTTP error.

Affecting anyone else?

Conan

All mine uploaded with some intermittence yesterday (UK) but there was only 20. All uploads today have gone straight up.

I now have dozens that wont upload with the same error.

All other projects I am running are uploading and downloading, just MLC wont upload, I have received downloads.

Conan

I now have dozens that wont upload with the same error.

All other projects I am running are uploading and downloading, just MLC wont upload, I have received downloads.

Conan

I suggest you ask your question on the Discord server

Have joined and asked on the Discord channel, still no solution yet and now getting a "server error feeder not running error" here at MLC.

I have 2 to 3 pages of results to return, I would prefer not to lose them,

Conan

Server error feeder and [mlc] disk full happened today but it appears to be up and running again. I had a load of tasks waiting upload on one computer but another computer uploaded fine, it appeared to be boinc-client holding off to some time in the future, I rebooted boinc-client and all but one task uploaded I suspect that one will upload within the next 24 hours so I'm not going to force it.

Boinc-client wouldn't download any new tasks because there were "too many uploads waiting" which is why I rebooted to force the issue.

Still no go for me.

All WUs (60 of them) fail with the "Transient HTTP error".

It has been over 3 days now and the way it is going I may lose the lot, which wont make me very happy.

Ah well we will see what happens, I have to get work done on my computer today so will try again later today.

I don't believe it is computer related as no other project I am running is having the same issue.

Conan

OK even though my new computer is only 3 weeks old it has been rebuilt (with new motherboard (different make), replaced the CPU (same type), replaced the GPU (same type), replaced the NVMe drive, replaced the power supply and the SATA leads).

All this has not helped me upload anything to MLC, but so far the computer after 3 hours is still on, I was having major crashes and I hope this fixes it all.

So I still have 60 work units to upload, but I can't contact the upload server due to this "Transient HTTP error".
If I do an "Update" of the project via the Boinc Manager I do contact the project as indicated on my account page.

I run multiple projects on this computer and none of them are having any issue uploading or downloading work.

Conan

Another day and still no upload of any one of my work units, all 60 still have the "Transient HTTP error" still.

Adding a "http debug" option to the Event Log shows

"OpenSSL SSL_read: error:0A000126:SSL routines::unexpected eof while reading, errno 0"
"info: Closing connection 769"
"HTTP error: Failure when receiving data from the peer"

Not sure how to read that but anything with an error in it is not a good thing.

Conan

Well I have worked it out and solved my issue.

For some reason after working for weeks MLC and later Ramunijan Machine stopped talking to their respective projects, I have no idea why.
I started checking anything that could stop uploads and downloads, being Linux, antivirus is limited but the Firewall was not blocking anything.
Checked cc_config.xml a number of times changing a few things but nothing worked.

Finally found an old Boinc forum posting from 2016 that questioned why a user was enforcing http 1.0 for file transfers and not letting Boinc use http 1.1.
Found that setting in my cc_config.xml, so I disabled it (changed from <http_1_0>1</http_1_0> to <http_1_0>0</http_1_0>), read the config files and uploads started working.

Also I can talk to Ramunijan Machine now as well.

Conan