Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Add timeout to SSH queue scan command #191

Merged
merged 11 commits into from
Jun 18, 2024
Merged

Add timeout to SSH queue scan command #191

Show file tree
Hide file tree
Changes from 10 commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
e2a5ba8
Add timeout to SSH queue scan command
Icemole May 31, 2024
213c745
Chagne Popen/communicate by run
Icemole May 31, 2024
bceb3e2
Add batch mode option to ssh command
Icemole May 31, 2024
3ca38b4
Remove stdout/err pipe
Icemole May 31, 2024
9880e42
Better formatting
Icemole May 31, 2024
d639048
Capture output on subprocess
Icemole May 31, 2024
7679653
Apply fix for AWS and load sharing engines
Icemole May 31, 2024
60fe641
Input instead of stdin
Icemole May 31, 2024
2b67c81
Revert timeout of subprocess to 30 seconds
Icemole Jun 3, 2024
f5fdfc7
Catch `subprocess.TimeoutExpired`
Icemole Jun 13, 2024
a94365a
Properly return when timeout is expired
Icemole Jun 18, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 8 additions & 6 deletions sisyphus/aws_batch_engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
""" This is an experimental implementation for the aws batch engine.
"""This is an experimental implementation for the aws batch engine.

WARNING: After running some setups I can currently not recommend using aws batch with Sisyphus.
AWS parallelcluster (https://aws.amazon.com/blogs/opensource/aws-parallelcluster/) looks like a easy way how
Expand Down Expand Up @@ -88,10 +88,12 @@ def system_call(self, command, send_to_stdin=None):
system_command = command

logging.debug("shell_cmd: %s" % " ".join(system_command))
p = subprocess.Popen(system_command, stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
if send_to_stdin:
send_to_stdin = send_to_stdin.encode()
out, err = p.communicate(input=send_to_stdin, timeout=30)
try:
p = subprocess.run(system_command, input=send_to_stdin, capture_output=True, timeout=30)
except subprocess.TimeoutExpired:
logging.warning("Timeout expired for command: %s" % " ".join(system_command))

def fix_output(o):
"""
Expand All @@ -105,9 +107,9 @@ def fix_output(o):
assert False
return o[:-1]

out = fix_output(out)
err = fix_output(err)
retval = p.wait(timeout=30)
out = fix_output(p.stdout)
err = fix_output(p.stderr)
retval = p.returncode

return out, err, retval

Expand Down
12 changes: 7 additions & 5 deletions sisyphus/load_sharing_facility_engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,10 +56,12 @@ def system_call(self, command, send_to_stdin=None):
system_command = command

logging.debug("shell_cmd: %s" % " ".join(system_command))
p = subprocess.Popen(system_command, stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
if send_to_stdin:
send_to_stdin = send_to_stdin.encode()
out, err = p.communicate(input=send_to_stdin, timeout=30)
try:
p = subprocess.run(system_command, input=send_to_stdin, capture_output=True, timeout=30)
except subprocess.TimeoutExpired:
logging.warning("Timeout expired for command: %s" % " ".join(system_command))

def fix_output(o):
# split output and drop last empty line
Expand All @@ -69,9 +71,9 @@ def fix_output(o):
assert False
return o[:-1]

out = fix_output(out)
err = fix_output(err)
retval = p.wait(timeout=30)
out = fix_output(p.stdout)
err = fix_output(p.stderr)
retval = p.returncode

# Check for ssh error
err_ = []
Expand Down
16 changes: 10 additions & 6 deletions sisyphus/simple_linux_utility_for_resource_management_engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -86,16 +86,20 @@ def system_call(self, command, send_to_stdin=None):
"""
if self.gateway:
escaped_command = [shlex.quote(s) for s in command] # parameters need to be shell safe when sending via ssh
system_command = ["ssh", "-x", self.gateway] + [" ".join(["cd", os.getcwd(), "&&"] + escaped_command)]
system_command = ["ssh", "-x", self.gateway, "-o", "BatchMode=yes"] + [
" ".join(["cd", os.getcwd(), "&&"] + escaped_command)
]
else:
# no gateway given, skip ssh local
system_command = command

logging.debug("shell_cmd: %s" % " ".join(system_command))
p = subprocess.Popen(system_command, stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
if send_to_stdin:
send_to_stdin = send_to_stdin.encode()
out, err = p.communicate(input=send_to_stdin, timeout=30)
try:
p = subprocess.run(system_command, input=send_to_stdin, capture_output=True, timeout=30)
except subprocess.TimeoutExpired:
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we catch the subprocess.TimeoutExpired exception here and return, then the whole logic about retrying in submit_helper

sisyphus/sisyphus/simple_linux_utility_for_resource_management_engine.py

Lines 239 to 246 in 51db853

while True:
try:
out, err, retval = self.system_call(sbatch_call)
except subprocess.TimeoutExpired:
logging.warning(self._system_call_timeout_warn_msg(command))
time.sleep(gs.WAIT_PERIOD_SSH_TIMEOUT)
continue
break
is obsolet.
If this was intentional, then maybe also remove this logic? and gs.WAIT_PERIOD_SSH_TIMEOUT would be ignored then

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah so it is actually handled. That was my earlier question about this. So then I would not catch it. Unless there is good reason to change this old behavior about TimeoutExpired.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes... I hadn't noticed that it would be catched later in the call stack. See #196.

logging.warning("Timeout expired for command: %s" % " ".join(system_command))
Icemole marked this conversation as resolved.
Show resolved Hide resolved

def fix_output(o):
"""
Expand All @@ -109,9 +113,9 @@ def fix_output(o):
assert False
return o[:-1]

out = fix_output(out)
err = fix_output(err)
retval = p.wait(timeout=30)
out = fix_output(p.stdout)
err = fix_output(p.stderr)
retval = p.returncode

# Check for ssh error
err_ = []
Expand Down
16 changes: 10 additions & 6 deletions sisyphus/son_of_grid_engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -87,16 +87,20 @@ def system_call(self, command, send_to_stdin=None):
:rtype: list[bytes], list[bytes], int
"""
if self.gateway:
system_command = ["ssh", "-x", self.gateway] + [" ".join(["cd", os.getcwd(), "&&"] + command)]
system_command = ["ssh", "-x", self.gateway, "-o", "BatchMode=yes"] + [
" ".join(["cd", os.getcwd(), "&&"] + command)
]
else:
# no gateway given, skip ssh local
system_command = command

logging.debug("shell_cmd: %s" % " ".join(system_command))
p = subprocess.Popen(system_command, stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
if send_to_stdin:
send_to_stdin = send_to_stdin.encode()
out, err = p.communicate(input=send_to_stdin, timeout=30)
try:
p = subprocess.run(system_command, input=send_to_stdin, capture_output=True, timeout=30)
except subprocess.TimeoutExpired:
logging.warning("Timeout expired for command: %s" % " ".join(system_command))

def fix_output(o):
"""
Expand All @@ -110,9 +114,9 @@ def fix_output(o):
assert False
return o[:-1]

out = fix_output(out)
err = fix_output(err)
retval = p.wait(timeout=30)
out = fix_output(p.stdout)
err = fix_output(p.stderr)
retval = p.returncode

# Check for ssh error
err_ = []
Expand Down
Loading