| Author |
Message |
|---|
|
|
|
I only can upload PPS only but can't upload genefer_extreme
"PrimeGrid | Temporarily failed upload of genefer_extreme_67902633_1_r477854148_0: transient HTTP error"
It because XGFN server is down...?But i keep get the new task....?
Any one can help?
Thank You
|
|
|
|
|
|
i also have this problem. did you recently update your nvidia driver to 526.98?
no sure if this is related, but experiencing the same problems as you. see the other recent post in this board. |
|
|
Yves Gallot Volunteer developer
Project scientist
Send message
Joined: 19 Aug 12
Posts: 843
ID: 164101
Credit: 306,522,117
RAC: 5,375
|
|
genefer v4.01 generates a "proof". The size of the file is 128MB for DYFL. This file was not uploaded or an error occured during data transfer. It's a network issue. |
|
|
|
|
|
not on my end i'd say. other projects upload fine, including non genefer primegrid tasks. |
|
|
|
|
i also have this problem. did you recently update your nvidia driver to 526.98?
no sure if this is related, but experiencing the same problems as you. see the other recent post in this board.
Yer i am using 526.98.....but...
updating to 527.37 did not solve the problem.
genefer tasks still get stuck at 0.06 MB upload.
pps uploaded just fine
But other people seem can't upload genefer task but pps is ok?
May be it just due to the file size...?GFN have large file size....
|
|
|
|
|
|
12/8/2022 8 AM | PrimeGrid | Temporarily failed upload of genefer19_63406395_0_r1494780881_0: transient HTTP error
12/8/2022 8 AM | | Internet access OK - project servers may be temporarily down.
12/8/2022 8 AM | PrimeGrid | Sending scheduler request: Requested by user.
12/8/2022 8 AM | PrimeGrid | Not requesting tasks: too many uploads in progress
Any one can help/____\
Can't get new job now
|
|
|
|
|
|
i'm getting similar errors.
12/8/2022 9:01:02 AM | | Project communication failed: attempting access to reference site
12/8/2022 9:01:02 AM | PrimeGrid | Temporarily failed upload of genefer_extreme_67902709_1_r736839561_0: transient HTTP error
12/8/2022 9:01:02 AM | PrimeGrid | Backing off 03:59:16 on upload of genefer_extreme_67902709_1_r736839561_0
12/8/2022 9:01:03 AM | | Internet access OK - project servers may be temporarily down.
i've suspended primegrid for now because tasks are expiring and that is a waste of energy |
|
|
 pschoeferVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 05
Posts: 686
ID: 845
Credit: 3,010,105,939
RAC: 630,276
                              
|
|
Just a random guess, but it has happened in the past: Are you using any antivirus software that might not like the large uploads?
____________
  |
|
|
|
|
|
It's a standard windows machine. I did just disable firewall and virus protection to try it but still no luck.
I also tried tethering through my mobile phone for a different internet connection. No luck.
Even the 2MB files fail. I now have 6 uploads in 2-128 MB range that keep failing to upload at 0.06MB. When I show up in the morning they sometimes have gone further than 0.06MB but never successful.
|
|
|
|
|
|
there's definitely something fishy going on here.
when i look at my 9 tasks which are still trying to upload (most of the times get stuck at 0.06 kb) i can see that 6/9 have another user who also worked on them. all of those have either aborted the workunit of have produced an error. |
|
|
streamVolunteer moderator
Project administrator
Volunteer developer
Volunteer tester
Send message
Joined: 1 Mar 14
Posts: 1051
ID: 301928
Credit: 563,881,725
RAC: 1,288
                   
|
there's definitely something fishy going on here.
when i look at my 9 tasks which are still trying to upload (most of the times get stuck at 0.06 kb) i can see that 6/9 have another user who also worked on them. all of those have either aborted the workunit of have produced an error.
I found these tasks. You client reported them as "successfully completed", although big file wasn't uploaded. So they failed validation. It's definitely not good, may be it's a bug in client.
Unfortunately, I cannot help you with upload problem. It happened sometimes and their reason always was a misconfigured router / equipment somewhere on the way between user and PG servers. Sometimes it was firewall / antivirus on user side, sometimes the problem was at higher level in internet service provider equipment. In later case, calling an ISP tech support and describing the problem ("cannot upload big file to server ....") helped sometimes. But these types of problem are difficult to diagnose because the glitch may affect only specific servers - for example, when Internet traffic passes through specific cross-continent or cross-country links.
|
|
|
streamVolunteer moderator
Project administrator
Volunteer developer
Volunteer tester
Send message
Joined: 1 Mar 14
Posts: 1051
ID: 301928
Credit: 563,881,725
RAC: 1,288
|
|
Important node: your computer is connecting to PG through IPV6 protocol. It's not so widespread and a chance that something is wrongly configured somewhere on the way is significantly higher. If possible, try to disable IPV6 in Windows network settings or give IPV4 higher priority than IPV6 - https://kb.firedaemon.com/support/solutions/articles/4000160803-prioritising-ipv4-over-ipv6-on-windows-10-and-11
All what I see on our side is a timeout waiting for data from your computer:
[Sun Dec 11 08:48:24.827063 2022] [fcgid:warn] [pid 22481] (70007)The timeout specified has expired: [client 2600:****:****:****:****:****:****:****:55692] mod_fcgid: can't get data from http client |
|
|
|
|
|
you are correct. disabling ipv6 on the windows host instantly allowed the upload to go through. too bad i already aborted most tasks today.
i haven't changed anything recently to these settings. maybe my isp changed something but i think i always had a public ipv6 address. this is pretty standard here.
thanks for solving this, i hope it helps others.
and i hope ipv6 will be supported at some point. |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 861
ID: 18447
Credit: 873,063,625
RAC: 1,344,899
         
|
|
The most common issue with ipv6 are MTU mismatches combined with ICMP Packet Too Big responses being blocked, which breaks Path MTU Discovery.
IMHO A better workaround than disabling ipv6 is to reduce the TCP MSS to a smaller value e.g. 1200.
But the real fix is that the ISP needs to get it's act together - it's almost 2023 and many people have this working just fine.
____________
 |
|
|
mikey
Send message
Joined: 17 Mar 09
Posts: 1895
ID: 37043
Credit: 825,161,243
RAC: 574,098
           
|
The most common issue with ipv6 are MTU mismatches combined with ICMP Packet Too Big responses being blocked, which breaks Path MTU Discovery.
IMHO A better workaround than disabling ipv6 is to reduce the TCP MSS to a smaller value e.g. 1200.
But the real fix is that the ISP needs to get it's act together - it's almost 2023 and many people have this working just fine.
Is that TCP MSS setting in the router? If so some of us using ISP provided routers can't change that setting. My old neighbors son said lowering the 'packet size' because the ISP doesn't support 'large packets' can help too but again that's in the router settings. |
|
|
|
|
|
i can't find out what TCP MSS is. i tried setting the MTU on windows host to 1200 but then it fails again.
|
|
|
mikey
Send message
Joined: 17 Mar 09
Posts: 1895
ID: 37043
Credit: 825,161,243
RAC: 574,098
|
i can't find out what TCP MSS is. i tried setting the MTU on windows host to 1200 but then it fails again.
Try this page https://community.fortinet.com/t5/FortiGate/Technical-Tip-Setting-TCP-MSS-value/ta-p/194518
at the bottom of the page it says this:
For MS Windows OS client enter:
ping -f -l 1472 <server ip>
For Apple MAC OS client enter:
ping -D -s <size> <server_ip>
-- For 1500 bytes MTU, 1500-28 (20 bytes=IP header and 8 bytes=ICMP header) = 1472
If the packet needs to be fragmented but the DF bit set, it means that the MTU is less than 1500 bytes.
Suppose it is possible to ping with a size of 1400 but not with a size of 1401, this means that the MTU value is 1400+28=1428 bytes.
The solution is either to fix the MTU in the network device or to set the TCP MSS in the firewall policy to 1428-40=1388 bytes.
my server ip here is 192.168.0.1 so I put that instead of the <server_ip> and it came back with 1472 like it says it would
The problem for my ISP is they add a beginning and end to each packet so sending a 1472 byte file is never going to fit into the max of 1500 bytes so it splits every request into TWO sends which is the problem for my slow downs
At my old house I changed the setting in the router so it applied to every pc not just the one I'm on right now but I can't get into the ISP's router settings to make that change so I may have to do it for every pc. |
|
|
