High Amount of Upload Failures?

[SpyShadow]

I have notice a increase of upload failures lately. Should I be concern? I have not made any changes lately.

[Mark]

I think I’m seeing the same thing for about the last hour. Mostly coming from satellite:12L9ZFwhzVpuEKMUNUqkaTLGzwY9G24tbiigLiXpmZWKwmcNDDs
========== UPLOAD ============
Successful: 6
Rejected: 0
Failed: 217
Acceptance Rate: 100.000%
Success Rate: 2.691%

[SpyShadow]

Lots of them

[Cmdrd]

Significant issues with file retrieval

I assume you mean on the storagenode side? Yes, there is a correlation. I believe before the update 95 uploads were started and after 80 were done the remaining 15 were cancelled. Now 110 are started and 30 of them cancelled. So that’s already a doubling of the context cancelled amount. But for some nodes the impact may be much higher. If you consistently came in as around the 20th to last node to finish the transfer, you would have succeeded with the previous settings but you fail with the ne…

Changes in the number of storage nodes being uploaded to and the ratio winning the race has changed with the latest update, a lot of people are seeing the same thing.

[Mark]

I also read in another thread that stress tester Stefan Benten is moving to another satellite. Perhaps he is operating from the europe-west satellite now?

[SpyShadow]

Should I just ignore it then?

[Mark]

If your node is otherwise working properly, than yeah, not much you can do about the context cancelled uploads.

[SpyShadow]

Guess seeing a bunch of those error messages had me concern. So why I asked about it here. I think those type of messages should be labeled as something else like “Canceled” vs just “Failed”.

[Miner]

My nodes started bringing up alot of those messages as well.
I could fix this by reducing the cpu load other processes on the servers use.
After lowering the physical cpu core usage of other processes, the “upload failed” messages from storj disappeared almost completely.

I hope this helps someone.

[ccristian2003]

I noticed the traffic pattern changed in February, more Ingress than Egress and also in the logs lots of upload canceled contexts. I guess we are still in testing phase, otherwise the node seem to work ok.

[Pentium100]

Basically, the less traffic there is, there is more competition among nodes and you will see more failures.

[Storgeez]

Again nobody is mentioning this: Context cancelled on all uploads/downloads.

[anon27637763]

Unfortunately…

In order to actually know if a particular data piece was successfully caught, one would need to query the logs for reportedly failed pieces and then query the sqlite DB … this is time consuming, resource consuming, as well as slightly risky since the sqlite DB is actively opened several times by the Storjnode software.

[Storgeez]

Uplink can simply send a “success” packet or something similar. As they say, they will fix this.