Suddenly getting many validate errors on S3 since yesterday:
It probably just means that the validator needs tweaking for the conditions of the new S3 tasks series. My guess is that adjustments will probably be made and the tasks then be resubmitted for validation - or something like that.
Teething problems like this have happened before.
As I mentioned earlier on, this transition seems quite unusual and quite rushed, so it's probably not a great surprise if there are some issues. I'm sure it will all get fixed in the end :-).
Any updates on the S3 validator?
Tom M
A Proud member of the O.F.A. (Old Farts Association). Be well, do good work, and keep in touch.® (Garrison Keillor) I want some more patience. RIGHT NOW!
Previous "validate errors" of O2MDFS3 tasks have been reset, validator is running. Will take a while to catch up and gather some statistics. There will be no assimilator yet, so still time to tune things in the validator.
at the time of writing, the status of S3 validation could easily be deduced by looking at the massive backlog of S3 tasks still waiting for validation. and the fact that the line item "O2MDFS3 validator" was not in existence. that is to say, the problem still wasn't fixed yet.
Previous "validate errors" of O2MDFS3 tasks have been reset, validator is running. Will take a while to catch up and gather some statistics. There will be no assimilator yet, so still time to tune things in the validator.
Thanks for the update. Many of my “validate error” tasks have already been sent out to other computers to try again. Is this necessary? I turned off GW temporarily to avoid picking up new copies of some of these “validate error” tasks that are being reassigned, but maybe I overreacted?
There will be no assimilator yet, so still time to tune things in the validator.
Sad that it seems to still need a tuning
I'm not sure what "it" is referring to in your message. You have a quote referring to the assimilator and the assimilator shouldn't need "tuning" since it just 'assimilates' validated results into more permanent storage. Once that happens, it's much harder to 'fix' validator issues. So it's quite normal NOT to assimilate until the validator is known to be performing as intended.
On the other hand, if "it" refers to the validator, how can it be "sad" if the Devs are taking extra precautions to make sure the validator isn't wrongly invalidating otherwise 'correct' results or allowing through 'incorrect' results?
Gary Roberts
)
Any updates on the S3 validator?
Tom M
A Proud member of the O.F.A. (Old Farts Association). Be well, do good work, and keep in touch.® (Garrison Keillor) I want some more patience. RIGHT NOW!
Tom M wrote: Any updates on
)
https://einsteinathome.org/server_status.php
_________________________________________________________________________
Sorry, there's still
)
Sorry, there's still something wrong with it. Had to postpone the work to fix some issues caused by the upload problems. I'm on it.
BM
Previous "validate errors" of
)
Previous "validate errors" of O2MDFS3 tasks have been reset, validator is running. Will take a while to catch up and gather some statistics. There will be no assimilator yet, so still time to tune things in the validator.
BM
Ian&Steve C. wrote:Tom M
)
https ... ?
Does not explain the status of the "validate error" that I have (over 300 at present for S3).
So how is the "Server Status" page going to help us answer the basic question?
Am I missing something?
Cheers!
Edit:
Just read Bernds messages - patience - patience,
but my above reply still holds at the time the original post from "Tom M" was written!
at the time of writing, the
)
at the time of writing, the status of S3 validation could easily be deduced by looking at the massive backlog of S3 tasks still waiting for validation. and the fact that the line item "O2MDFS3 validator" was not in existence. that is to say, the problem still wasn't fixed yet.
_________________________________________________________________________
... very well deduced
)
... very well deduced ...
THANKS
Bernd Machenschalk
)
Thanks for the update. Many of my “validate error” tasks have already been sent out to other computers to try again. Is this necessary? I turned off GW temporarily to avoid picking up new copies of some of these “validate error” tasks that are being reassigned, but maybe I overreacted?
Bernd Machenschalk
)
Sad that it seems to still need a tuning
Highlander wrote: Bernd
)
I'm not sure what "it" is referring to in your message. You have a quote referring to the assimilator and the assimilator shouldn't need "tuning" since it just 'assimilates' validated results into more permanent storage. Once that happens, it's much harder to 'fix' validator issues. So it's quite normal NOT to assimilate until the validator is known to be performing as intended.
On the other hand, if "it" refers to the validator, how can it be "sad" if the Devs are taking extra precautions to make sure the validator isn't wrongly invalidating otherwise 'correct' results or allowing through 'incorrect' results?
Cheers,
Gary.