-
Notifications
You must be signed in to change notification settings - Fork 192
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
A way to monitor/watch tails of output files that being written on remote #6666
Comments
I'm strongly against the second approach, as that would introduce an entirely new To avoid confusion, we could also modify the flag to, e.g., Only thing we need to be aware about is opening many transport connections/keeping one alive all the time, but probably that won't really be a problem. |
this log files can be very long thousands of lines.. |
I agree not adding a top level command. Indeed, outputcat at the moment means "retrieved files". We can have a clear flag, but then with the --remote option, you are really looking at a different node. If we think this makes things easier that's OK, but I fear that it can make things even more confusing. Also, I think we are mixing up getting files from the remote (as a general task, can be run also after calc ends, the endpoint above would work) with efficient monitoring while the job is running (we should then add a flag for a "tail -f" functionality, --watch would indeed be a good name, but we can brainstorm a bit more - BTW, what does the -f option of tail stands for?). Probably indeed, since give files can be big and network transfer is slow, having specific --head/--tail options for this command would be nice (but not useful for the standard outputcat command, I guess? One more reason to keep them separated?) and can be combined with --tail if one wants to monitor but just starting from the last lines, rather than printing the whole file and then continue monitoring the file). Final note, the --head/--tail command could (should?) have an option to select how many first/last lines to show (if feasible to implement easily) |
Hi @giovannipizzi , |
One way to do it, is something like
verdi calcjob outputcat <pk> <Relative_PATH> --watch
that would tail the latest linesbut that might be confusing for the users, because this is known to users as a command that prints out the retrieved files, not the remote ones.
A better suggestion might be
verdi watch <pk> <Relative_PATH>
verdi watch -n <seconds> <pk> <Relative_PATH> --tail <n_lines>
@agoscinski, @GeigerJ2, @Technici4n, @giovannipizzi
I vote for the second approach.
The text was updated successfully, but these errors were encountered: