Previously I’ve mapped the drives in windows and put credentials in there.
After a reboot those drives are not reconnected by windows if not used (red cross)
so I let deadline remap those drives to make connection again (without credentials)
how even, now when I let deadline map those drives it completely removes them from the slaves.
resulting in renders failing.
The drives are on a Synology NAS and the slaves are win 7 machines.
what to do here?
if I do put in credentials it will destroy the mapped drives on my work station (those are already mapped with other credentials)
If the option to only map if unmapped is enabled, then Deadline won’t map the drive if it already exists. If that option is disabled, Deadline will try to remap each time. However, Deadline never explicitly unmaps a network drive, so I’m not exactly sure what would be causing that behavior.
It would help to see the logs from the slaves that are trying to map the drives to see if there is any useful information.
mmh… today after resetting the repository and starting fresh I again configured the mapped drives to always be mapped.
This, again, removed the to-be-mapped drives from my nodes.
Still no clue how it can be, but figured to get back on it.
Originally the mapped drives have authentication, but I haven’t set this through deadline.
Perhaps this causes an issue / conflict in some weird way?
when I check the error logs I get this line for each location:
2014-08-09 02:16:52: 0: Error mapping Y: to //nas/Data/Resources/Assets: The network name cannot be found..
when I then login to the computer and run this line via a bat file
net use y: \\nas\Data\Resources
There is no problem at all and the drive is mapped.
Attached a video where there are 3 drives to be mapped.
-In the beginning you see my repository settings where the Z drive has been set to “Only Map if Unmapped”
-then I slide in the Slave teamviewer window where I just freshly re-mapped the drives again (deadline slave/launcher are off) and everything works
-The deadline launcher is started and the slave is running.
-When explorer is opened again the Z: drive is missing.
This is solved by setting the mapped drives to always map. But this only works if the drives are already mapped in the first place.
When the drives are not there already it doesn’t add the drives and gives the errors as above (sorry… no video of that )
It all seems a bit odd somehow. And I’m sure this is only since DL7 (still haven’t updated to the latest beta btw…)
–edit–
there is one node that doesn’t give issues.
It’s an older one that was previously a workstation.
All run the same windows 7 64bits SP1 so I checked the updates.
The working node misses the updates that are attached in the image.
Not sure if these are hardware related and the newer nodes don’t need these updates.
But figured might be useful to mention. MappedDriveDead.mp4 (971 KB)
We have not been able to reproduce the behavior that you are experiencing, following the steps you described and the video you shared. Maybe try updating windows on the offending slave machines and grabbing the latest DL7 beta and see if you are still experiencing problems with this.
)