I am trying to deploy a small database to our production server.
Last time I deployed this same database the whole process took 39 seconds.
Now it reached the fetching step and stucks there forever. I tried to let it wait up to 9 minutes before aborting the deployment.
What can be wrong?
Deploying on another server succeeded.
I tried to update OttoFMS on both servers to 4.9.0 with no success.
I tried to download the debug logs, but found nothing logged anywhere between the waiting and aborting steps:
|Waiting for upload to complete|fetching|info|22076|
|Deployment was aborted|unknown phase|error|124848|
I tried to update OttoFMS on both servers to 4.9.0 with no success.
Does that mean you couldn’t upgrade to OttoFMS to 4.9.0 or that upgrading didn’t help? If it didn’t help then I assume you tried it at least 2 times. Is that right?
If there is some issue with the network that is causing sending to be very slow, you could try reversing it from a PUSH to a PULL.
See this doc
Also I’d still like to see the debug logs. If you could DM them to me that would be great.
Trying some more tests…
Trying to use Otto to deploy a Test database there (install).
This is failing with some permissions error:
Build push error: tus: unexpected response while creating upload, originated from request (method: POST, url: http://edna.24u.cz/otto/uploads/files, response code: 403, response text: <!--#set var="TITLE" value="Access forbidden!" --><!--#include virtual="include/top.html" --> <!--#if expr="v('REDIRECT_URL') =~ m:/$:" --> You don't have permission to access the requested directory. There is either no index document or the directory is read-protected. <!--#else --> You don't have permission to access the requested object. It is either read-protected or not readable by the server. <!--#endif --> <!--#include virtual="include/bottom.html" --> , request id: n/a)
Tried to manually upload a file via File Browser to the OTTO Inbox. That succeeded.
That error looks like your servers cannot communicate from source to destination. You should be able to run the deployment if you turn off the push build as todd described above. Thank you!
Thanks guys, it seems that pull has succeeded. I am still currious though to find out why the push stopped working when it worked in the past, and also what I may be doing wrong with my test deployments… @toddgeist going to DM you the logs and any other info you find useful to examine.
The “Access Forbidden” text in the error you got is our clue for why its failing. Do you have IP filtering turned on or some sort of Firewall in place on the destination server?
Pull vs Push just comes down to if your network is open between the two servers in one direction or other other. Many things can get in the way of that. Firewall issues, DNS issues, SSL cert issues, or IP Filtering.
Also networks and other hardware config issues can dramatically impact the speed of transfer. So one direction maybe MUCH faster than the other.