Snapin unable to deploy
-
@Hanz the code operates as:
Check if the file exists and is readable. It is done via ftp methods so if FTP is not correct this could be presented as well.
Essentially, if either the file doesn’t exist or the file is not readable to the FTP user, it will NOT be able to send the file.
I’d recommend verifying that the nodes that are expected to have the files actually have them, and that their ftp information is correct.
-
@Hanz said All snapins are present, and permissions are correct…ftp shows “Index of /” (but nothing there)…Log Viewer is acting as though one of my other nodes is the server. With the Multicast, Scheduler information and so on. While the actual server still shows “no data to read”
-
@Hanz said:
…ftp shows “Index of /” (but nothing there)…
Doesn’t really sound like FTP is working the way it shoudl?!?
-
Yeah, Another node has assumed the role of Server after being set up as server first then /opt/fog/.fogsettings changed to accomodate Storage node install.
-
As I mentioned before in another thread relating to MYSQL error, All nodes under Log Viewer are showing results as though they’re servers. I’m confused as to why one node has decided to become the server…Where in the database is a node “declared” as the main server node ?
-
Master Nodes
-
@Tom-Elliott yes, all in their own groups, all show the following under Image/Snapin Replicator, but one node
“[10-23-15 11:15:39 am] * | This is not the master node” on everything except scheduler. My actual server shows “no data to read” -
What I was answering is the systems who “register as servers” are those who are master nodes.
-
My Server “Fog-Server” is typically listed on this page, and it is not…kernel shows, but that’s it.
-
@Tom-Elliott I was given an invalid random ftp password again from installer “/” character will error a lot of things out.
-
I don’t follow. The / should not error anything out. We are using single quote so there is no escaping of characters that should be happening.
-
Sorry, they’re all master nodes, for each school, in it’s own group…I can only assume that the passwords were correct on all machines. Im 99% sure all nodes were double checked by removing and re-adding to .fogsettings then re-installing, and then fog user passwd updated, then GUI node password verified. The only thing I did was replace the “/” in the password with “1” and doing the afforementioned and it all worked correctly, immediately.
I think it might be worth looking into, since User passwords are mitigated to the eligible characters (could be global)
-
This thread can probably be closed now…I appreciate your gentlemen’s help. I think the ftp wasn’t working correctly as @Uncle-Frank mentioned. Snapins are now functioning correctly.
-
Problem manually solved.
I also had a generated password with a “/”. When I tried to deploy a snapin, it failed with the same error @Hanz described.
A login with an an ftp-client redirected me into the users home directory. I changed “/” into a different character (passwd and GUI) and the ftp-client redirected me into /opt/fog/snapins.
Snapins now work without an error! -
@mp12 so we should not allow passwords with slashes?
-
@Wayne-Workman maybe someone can reproduce the error, by creating a password with a “/” to eliminate individual mistakes from my site?
-
@mp12 Do you think it could be possible that the password was simply set incorrectly somewhere and you resetting it fixed it? I think this is the most plausible reason why it was fixed.
-
@Wayne-Workman maybe. I’ll check this on monday.
-
@Wayne-Workman just replicated the error.
Reconfigured the password in GUI and passwd with a “/”.
–> “Failed to find snapin file”
Reconfigured the password in GUI and passwd without a “/”.
–> Works like a charm!
I tried with installed FogClient and with the FogClient Debugger.
-
@mp12 Thanks for testing, now we know. @Developers thoughts?
