Argument list too long: '/bin/sh'

[jgrund]

Relevant code:

https://github.com/ClusterLabs/crmsh/blob/d0150a8874d800caf7abb57c8f17e1a73f9dd0a6/crmsh/ui_node.py#L341

When trying to run crm node standby I see the following error:

[root@node7 ~]#  crm --debug node standby
2022/06/15 17:35:23
DEBUG: pacemaker version: [err: ][out: CRM Version: 2.1.0-8.el8 (7c3f660707)]
DEBUG: found pacemaker version: 2.1.0-8.el8
Traceback (most recent call last):
  File "/usr/lib/python3.6/site-packages/crmsh/ui_context.py", line 92, in run
    rv = self.execute_command() is not False
  File "/usr/lib/python3.6/site-packages/crmsh/ui_context.py", line 284, in execute_command
    rv = self.command_info.function(*arglist)
  File "/usr/lib/python3.6/site-packages/crmsh/ui_node.py", line 341, in do_standby
    rc = utils.diff_and_patch(xmlutil.xml_tostring(orig_cib), xmlutil.xml_tostring(cib))
  File "/usr/lib/python3.6/site-packages/crmsh/utils.py", line 3079, in diff_and_patch
    rc, cib_diff, err = get_stdout_stderr(cmd)
  File "/usr/lib/python3.6/site-packages/crmsh/utils.py", line 812, in get_stdout_stderr
    stderr=subprocess.PIPE)
  File "/usr/lib64/python3.6/subprocess.py", line 729, in __init__
    restore_signals, start_new_session)
  File "/usr/lib64/python3.6/subprocess.py", line 1364, in _execute_child
    raise child_exception_type(errno_num, err_msg, err_filename)
OSError: [Errno 7] Argument list too long: '/bin/sh'
ERROR: node.standby: [Errno 7] Argument list too long: '/bin/sh'

What appears to be happening is the cib being printed is too long and overflows MAX_ARG_STRLEN

[liangxin1300]

Hi @jgrund Don't know if https://github.com/ClusterLabs/crmsh/pull/984 works for you

Thanks!

[jgrund]

Looks like the argument issue is fixed. However, crm node standby returned a 0 exit code on a node but the node never moved into standby. This happened on 1/4 nodes tested. Not sure if it's related to this change or not.

[liangxin1300]

Looks like the argument issue is fixed. However, crm node standby returned a 0 exit code on a node but the node never moved into standby. This happened on 1/4 nodes tested. Not sure if it's related to this change or not.

Could you please run crm -d node standby <node> to see what's debug info printed when failed to standby?