I recently have a problem with password. When I type “cryosparcm start”, a small window popped out and request a password (the computer password), as shown in this screenshot. when I run or kill a job, this window come out too. There is another error report in another screenshot. Appreciate for a solution to remove this password request.
Is this a combined cryoSPARC master/worker instance on a single computer?
Has it worked previously for GPU-enabled cryoSPARC jobs without requesting a password?
Unless some stringent hostname matching requirements are met (which may fail due to computer or network configuration changes), password-less ssh access to worker nodes is required for cryoSPARC function.
Dear wtempel,
Thank you for your reply and comments.
This problem was happened right after re-configuration of the computer. The master/worker is on a single computer. The network configuration was also changed after reconfiguration of the computer due to unknown reason (we will test that). In the config.sh: Cryosparc_Force_Master_Hotsname = true. Do you need more information?
Thank you, Lan
I want to make a correct. The password issue does not apply for those jobs, like Select 2D classes…
Interactive jobs like Select 2D classes are designed to run directly on the master even in multi-host cryoSPARC instances and should not experience the problem you saw for non-interactive jobs.
Please can you post the output of these commands:
grep -iv license /path/to/cryosparc_master/config.shhost $(hostname -f)cryosparcm cli "get_scheduler_targets()[0]['hostname']"
The screenshot suggests some transcription errors (letters, bracketing, quoting).
Would it be possible to paste the command above into the terminal?
Traceback (most recent call last):
File “/home/cryosparc_user/software/cryosparc/cryosparc_master/deps/anaconda/envs/cryosparc_master_env/lib/python3.7/runpy.py”, line 193, in _run_module_as_main
“main”, mod_spec)
File “/home/cryosparc_user/software/cryosparc/cryosparc_master/deps/anaconda/envs/cryosparc_master_env/lib/python3.7/runpy.py”, line 85, in _run_code
exec(code, run_globals)
File “/home/cryosparc_user/software/cryosparc/cryosparc_master/cryosparc_compute/client.py”, line 92, in
print(eval(“cli.”+command))
File “”, line 1
cli.get_schedular_target({0} {hostname]
[cryosparc_user@r16763 /]$ cryosparcm cli “get_scheduler_targets()[0][‘hostname’]”
r16763
Hi, is this what you want to see? Lan
Thanks for confirming that.
I think the problem occurs due to a mismatch between
r16763.ttuhsc.edu (includes domain)
and
r16763 (does not include domain)
A workaround would be to let cryoSPARC ignore the domain name.
- stop cryoSPARC (will affect running jobs):
cryosparcm stop - make a backup copy of
/home/cryosparc_user/software/cryosparc/cryosparc_master/config.sh - append the following line to
/home/cryosparc_user/software/cryosparc/cryosparc_master/config.sh:
export CRYOSPARC_MASTER_HOSTNAME="r16763" - Confirm that the final line of the output of
tail -n 1 /home/cryosparc_user/software/cryosparc/cryosparc_master/config.sh
appears exactly as in the previous step (capitalization, no spaces around=, etc.) - start cryoSPARC:
cryosparcm start
Can you now submit jobs without a password?
[cryosparc_user@r16763 ~]$ cryosparcm start
ERROR: Re-run this command on the master node: r16763.
Alternatively, set CRYOSPARC_FORCE_HOSTNAME=true in cryosparc_master/config.sh to suppress this error.
If this error message is incorrect, set CRYOSPARC_HOSTNAME_CHECK to the correct hostname in cryosparc_master/config.sh.
I remember that I faced this error when I first run cryosparcm start. then I did CRYOSPARC_FORCE_HOSTNAME=true
Yes please try appending the line
export CRYOSPARC_FORCE_HOSTNAME=true
to /home/cryosparc_user/software/cryosparc/cryosparc_master/config.sh
and then
cryosparcm restart
I have this line at line 2.
Well I added this, i have the same error:
[cryosparc_user@r16763 ~]$ cryosparcm start
ERROR: Re-run this command on the master node: r16763.
Alternatively, set CRYOSPARC_FORCE_HOSTNAME=true in cryosparc_master/config.sh to suppress this error.
If this error message is incorrect, set CRYOSPARC_HOSTNAME_CHECK to the correct hostname in cryosparc_master/config.sh.
After remove the line export CRYOSPARC_MASTER_HOSTNAME="r16763" and the cryosparc can be started. But the open SSH window showed up before the test
CryoSPARC master started.
From this machine, access cryoSPARC at
http://localhost:39000
and access cryoSPARC Live at
http://localhost:39006
There seems to be a confusion about variable names.
The screenshot in your earlier post shows
CRYOSPARC_FORCE_MASTER_HOSTNAME
whereas the valid variable name is shorter:
CRYOSPARC_FORCE_HOSTNAME
Have you tested cryoSPARC with these simultaneous settings:
export CRYOSPARC_MASTER_HOSTNAME="r16763"
export CRYOSPARC_FORCE_HOSTNAME=true
(other necessary variables inside /home/cryosparc_user/software/cryosparc/cryosparc_master/config.sh not shown here)?
export CRYOSPARC_LICENSE_ID=“XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX”
export CRYOSPARC_DB_PATH="/home/cryosparc_user/software/cryosparc/cryosparc_database"
export CRYOSPARC_BASE_PORT=39000
export CRYOSPARC_DEVELOP=false
export CRYOSPARC_INSECURE=false
export CRYOSPARC_CLICK_WRAP=true
export CRYOSPARC_FORCE_MASTER_HOSTNAME=“r16763”
export CRYOSPARC_FORCE_HOSTNAME=true
this is the current version as you suggeste. the problem still exists. any input?
thanks, Lan
What happens if you replace this line with
export CRYOSPARC_MASTER_HOSTNAME="r16763"
and restart cryoSPARC?
Great. this change works now. No password request windown appeared up and started to run the job.
Thanks! after give feedback if I can do the presetup-and-run! Lan