zstash Globus functionality has become overly cumbersome

[golaz]

Request criteria

• I searched the zstash GitHub Discussions to find a similar question and didn't find it.
• I searched the zstash documentation.
• This issue does not match the other templates (i.e., it is not a bug report, documentation request, feature request, or a question.)

Issue description

Using zstash with Globus is now much more cumbersome than it used to be; to the point that it may become questionable whether integrating Globus actually adds value to zstash.

Here is the workflow I have to use:

(1) Login to Globus web interface and activate end points

(2) Delete existing globus cfg file

rm ~/.globus-native-apps.cfg

(3) Start interactive zstash test transfer, something like

zstash create -v --non-blocking --hpss=globus://nersc/home/g/golaz/E3SMv3.LR/tmp tmp

During the the interactive test run, zstash will ask to copy and paste authorization code twice. zstash runs but files are not transferred.

(4) Start a second interactive zstash transfer

zstash create -v --non-blocking --hpss=globus://nersc/home/g/golaz/E3SMv3.LR/tmp tmp

This one should complete without issue or any prompt.

(5) Start long transfer (in a screen session to keep it running)

zstash create -v --non-blocking --hpss=globus://nersc/home/g/golaz/E3SMv3.LR/${EXP} . 2>&1 | tee zstash/zstash_create_${stamp}.log

Note that this transfer is limited to 48 hours due to Globus token expiration. Given the relatively slow transfer speeds between chrysalis and NERSC HPSS (~100 MB/s), 48 hours is often insufficient for a large simulation.

(6) Using Globus web interface, manually transfer zstash files that were not transferred due to token expiration.

(7) Repeat steps (2) to (4) above. Restart zstash archiving that stopped:

zstash update -v --non-blocking --hpss=globus://nersc/home/g/golaz/E3SMv3.LR/${EXP} 2>&1 | tee zstash/zstash_create_${stamp}.log

I should note that in the not so distant past, only steps (1) and (5) were needed. It is also possible that I'm doing something wrong, in which case I'd like to know what.

[golaz]

Related to #338

[forsyth2]

Note that this transfer is limited to 48 hours due to Globus token expiration

@golaz @TonyB9000 Sasha suggested

 --deadline [%Y-%m-%d|%Y-%m-%dT%H:%M:%S|%Y-%m-%d %H:%M:%S]

    Set a deadline for this to be canceled if not completed by.

and to look at https://docs.globus.org/cli/reference/transfer/.

[forsyth2]

#380 resolves steps 2-4 above. See #380 (review) for details.

[TonyB9000]

@forsyth2 The "--deadline" looks interesting, but according to the Great AI this only limits the end-time and is independent of token expirations, credential expirations, etc. It cannot be used to "lengthen" a run, only to terminate. According to the AI, there is no built-in "upper limit" to how long a transfer can run, it the tokens are auto-refreshed.