-
Notifications
You must be signed in to change notification settings - Fork 4.9k
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
Fix closing sessions #6114
Fix closing sessions #6114
Conversation
openhands/server/session/manager.py
Outdated
if sid in self._detached_conversations: | ||
conversation, _ = self._detached_conversations.pop(sid) | ||
self._active_conversations[sid] = (conversation, 1) | ||
logger.info(f'Reusing detached conversation {sid}') | ||
return conversation |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
why did we lose this?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I guess we just leave _attached_conversations
until the whole thing closes? That seems reasonable actually...
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The concept of stored detached conversations was replaced with a general concept of session staleness. A session is considered stale and subject to close if...
- It does not have any connections to it.
AND... - It has not had an update within the close_delay (Now 15 seconds by default).
Note: I think there may actually have been a bug here before my changes where the stale check was initialized along with the runloop and was not always being hit.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I assume you mean 15 minutes? 😅 15 seconds seems unbelievably low, just a quick tab away
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Correct. It is 15 minutes. (I actually changed this from 15 seconds to 15 minutes on Monday:
close_delay: int = 900 |
openhands/server/session/manager.py
Outdated
sids = {sid for sid, _ in items} | ||
return sids | ||
|
||
async def get_running_agent_loops_in_cluster( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
async def get_running_agent_loops_in_cluster( | |
async def get_running_agent_loops_remotely( |
this seems like maybe a better name?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done!
openhands/server/session/manager.py
Outdated
logger.info( | ||
f'Attached conversations: {len(self._active_conversations)}' | ||
) | ||
logger.info( | ||
f'Detached conversations: {len(self._detached_conversations)}' | ||
) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
why remove?
openhands/server/session/manager.py
Outdated
if sid in self._detached_conversations: | ||
conversation, _ = self._detached_conversations.pop(sid) | ||
self._active_conversations[sid] = (conversation, 1) | ||
logger.info(f'Reusing detached conversation {sid}') | ||
return conversation |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I guess we just leave _attached_conversations
until the whole thing closes? That seems reasonable actually...
async def _cleanup_session_later(self, sid: str): | ||
# Once there have been no connections to a session for a reasonable period, we close it | ||
try: | ||
await asyncio.sleep(self.config.sandbox.close_delay) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We need to remove this config right?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
AFAIK, there are OSS users that use this value - they have a use case where they want a session to persist for 8 hours while there is no connection to it. (As opposed to the 15 seconds we have by default)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yep, we have been using a long N hours close_delay
to keep our workspaces running around even after every browser closes.
With this new PR, is there a better way to achieve the same effect?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@diwu-sf - The settings you currently use should be fine - but you may get away with a shorter delay because the new behavior is that a conversation will be stopped if all three of the following are true:
- It has not been updated in
close_delay
seconds. - There are no connections to it.
- The agent is not in a running state. (This one is new!)
Now that I think about it, one thing that may affect you is that we have introduced a limit of 3 concurrent conversations per user. (So if you already have 3 running and start another it will kill one of the old ones regardless of the 3 criteria above - this is designed to stop the system crashing due to users trying to start too many concurrent docker containers). If this will affect you, we can introduce a config setting for this too.
…nds into fix-closing-sessions
We return the state, ERROR, or None
@@ -111,14 +111,11 @@ async def start( | |||
) | |||
self._initializing = False | |||
|
|||
def close(self): | |||
async def close(self): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've been trying to get rid of async close methods. Not sure if that's a goal worth pursuing, doesn't have to block this PR
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The reason this is now async is that it does send a final message down any connected socket indicating that the session is closing. (This is so that if a user deletes a conversation to which they are connected they get an appropriate message)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
FWIW, not about agent session, but it's generally a goal worth pursuing where possible IMHO. Timing can make things more complex/fragile in the execution of a multi-agent run, if some events may in theory come after their controller is closed or viceversa.
@@ -16,7 +17,7 @@ | |||
from openhands.runtime.base import Runtime | |||
from openhands.security import SecurityAnalyzer, options | |||
from openhands.storage.files import FileStore | |||
from openhands.utils.async_utils import call_async_from_sync, call_sync_from_async | |||
from openhands.utils.async_utils import call_sync_from_async | |||
from openhands.utils.shutdown_listener import should_continue | |||
|
|||
WAIT_TIME_BEFORE_CLOSE = 300 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm inclined to reduce this to something like 30
, which would make problems more apparent and easier to debug. Any concerns with that?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The remote runtime can occasionally take more than 30 seconds to start for me - I'll reduce it to 90. for now, and we can revisit later.
controller = self.controller | ||
if controller: | ||
return controller.state.agent_state | ||
if time.time() > self._started_at + WAIT_TIME_BEFORE_CLOSE: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If you take my comment above this probably needs to change
) | ||
running_sids.union(running_cluster_sids) | ||
return running_sids | ||
logger.warning(f'error_cleaning_stale: {str(e)}') |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is a very large block to have a blanket exception catch. It worries me a bit. This should probably be error
at least since it's unexpected
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Catching Exception
instead of error
will mean it does not catch things like KeyboardInterrupt
- this is really just to make sure that we don't stop cleaning up stale conversations due to an unexpected error
Comments here are non-blocking. Let's just make sure the latest commit has been thoroughly tested (especially in a multi-replica mode, and especially with multiple users connected) before merging |
…nds into fix-closing-sessions
Looks like this cause e.g., This commit adds new testcases that relies on CI error: https://github.com/All-Hands-AI/OpenHands/actions/runs/12793391296/job/35666481217 I just locally reproduced this, it seems after reverting this PR, the test started to work. With this commits, the zip downloaded from runtime has a size of |
This reverts commit 8795ee6.
This reverts commit 8795ee6.
This reverts commit 8795ee6.
End-user friendly description of the problem this fixes or functionality that this introduces
This PR improves the handling of multiple conversations and session management in OpenHands. It ensures that user workspaces are preserved even after disconnections or server restarts, and implements a smart session management system that automatically handles conversation limits.
Improved multi-conversation support with automatic session management and workspace preservation. Users can now maintain multiple conversations across different tabs while ensuring their work is preserved, even after disconnections or server restarts.
Summary of Changes
Acceptance Criteria for Multi-conversation Runtime Management
Recovery
Conversation Limits
Testing Instructions
To run this PR locally, use the following command: