68 Commits (7d211d431e5bea69347c19f64e5295df93dd5657)

Author	SHA1	Message	Date
Kevin Bates	2d688f2071	Prevent access through 'NoneType' when closing activity_stream ... Although I'm unable to reproduce the issue, its a safe change to prevent an AttributeError from occuring ('NoneType' object has no attribute 'close'). The user that reported this is attempting to launch a kernel and I believe the launch only partially completed such that `kernel._activity_stream` did not get established. (This occurred from Jupyter Enterprise Gateway where we deal with remote kernel launches across resource-managed clusters, so things are a bit more involved relative to kernel establishment.)	8 years ago
Min RK	a53d2c8489	flush iopub before sending restarting/dead status messages ... to ensure (approximately) that all messages from the stopped kernel are delivered before the dead/restarting message	8 years ago
M Pacer	c6dd032bf8	add kernel_info_timeout traitlet to wait for slow kernel startups ... This affects both the MappingKernelManager and the ZMQChannelsHandler(by extension). This allows one setting to apply to both startup andrestarting.	8 years ago
Kevin Bates	550120906c	Fix leak of iopub object in activity monitoring ... After analyzing various leaked items when running either Notebook or Jupyter Kernel Gateway, one item that recurred across each kernel startup and shutdown sequence was the a zmq socket instance associated with the iopub port. The leaked instance occurs because the activity monitor and normal port creation logic both call `create_iopub`. Although this ens up creating a _wrapper_ around the same port, the object (i.e., wrapper) created by the activity monitor is leaked. By setting the kernel manager's `_activity_stream` member to `None` when the kernel is shutdown, cleanup of the iopub wrapper object will take place.	8 years ago
Min RK	f77e3435cf	tornado 5 fixes in tests ... - start asyncio loop explicitly in notebook thread (tornado 4 starts per-thread loops automatically, asyncio doesn’t) - remove unsupported ‘io_loop=‘ arg in websocket test application runs fine with tornado 5, just a few test bits needed updating	8 years ago
Min RK	a2f72da17f	Merge pull request #2963 from takluyver/shutdown-no-kernels ... Config option to shut down server after N seconds with no kernels	8 years ago
Min RK	1deb0aec16	tornado 5: PeriodicCallback loop arg will be removed (#3034) ... * tornado 5: PeriodicCallback loop arg will be removed PCs are always run with the current eventloop, which is what the explicitly passed loop always is for us already * Don't double-close socket & stream closing stream closes the socket * remove now-inaccurate comment	8 years ago
Thomas Kluyver	bfe012e507	Merge pull request #2910 from mheilman/cull_idle_timeout_minimum ... change cull_idle_timeout_minimum to 1 from 300	8 years ago
Thomas Kluyver	9ec57bfdf5	Config option to shut down server after N seconds with no kernels	8 years ago
Michael Heilman	b31194dfe3	removed extraneous cull_idle_timeout_minimum	8 years ago
Michael Heilman	f8ef573592	added warning about culling kernels with poor network connections	8 years ago
Min RK	e4186c47c0	allow disabling offline message buffering ... escape hatch if the new buffering is causing anyone trouble	8 years ago
Michael Heilman	241c7e162c	change cull_idle_timeout_minimum to 1 from 500	8 years ago
Kyle Kelley	43a97807fc	buffer messages when websocket connection is interrupted (#2871) ... * provide some top level comments * implement buffering of messages on last dropped connection - buffer is per-kernel - session_key is stored because only a single session can resume the buffer and we can't be sure - on any new connection to a kernel, buffer is flushed. If session_key matches, it is replayed. Otherwise, it is discarded. - buffer is an unbounded list for now * restore actual zmq channels when resuming connection rather than establishing new connections fixes failure to resume shell channel * hookup restart callbacks in open instead of in `create_stream`, which is not called on reconnect * improve handling of restored connections in js - dismiss 'connection lost' dialog on reconnect - set busy status on reconnect (if not busy, idle will come soon after via kernel_ready)	8 years ago
Min RK	962c5ccd80	stop using `@json_handlers`	9 years ago
dhirschf	6bb89b717a	Workaround for Windows Containers ... On Windows `os.stat` treats host mapped volumes as broken symlinks	9 years ago
Peter Parente	c8ba55da21	Do not cull connected kernels by default	9 years ago
Peter Parente	85ba2bdabd	Fix help for cull_connected and cull_busy	9 years ago
Peter Parente	d9b4c99180	Do not cull kernels doing work by default	9 years ago
Peter Parente	1550a04848	Add cull_busy, cull_connected options ... Additional logic to support culling or skipping kernels that are busy or that have connections.	9 years ago
Min RK	f4d2c246d5	use dedicated Session for activity monitor ... instead of re-using kernel.session, which can cause hash collisions if other objects use `kernel.session`.	9 years ago
Thomas Kluyver	05979d9a4b	Merge pull request #2368 from minrk/stream-data-rate-boogaloo ... only consider stream outputs for data-rate limit	9 years ago
Min RK	3d7c9b51a6	only consider stream outputs for data-rate limit	9 years ago
Min RK	15341ac758	Revert "only consider stream output for data rate limit"	9 years ago
Min RK	59c9df1b89	show current rate limit values in rate limit messages	9 years ago
Min RK	5da6eb6a41	only consider stream output for data rate limit ... large stream outputs cause much more problems than image output this does open up to large HTML and/or displayed text output, but those seem to behave well more often than they don't.	9 years ago
Kevin Bates	ff19c4cd7c	enforce minimum timeout, ensure exceptions don't prevent culling of others	9 years ago
Kevin Bates	c3f753faf8	incorporate review recommendations	9 years ago
Kevin Bates	2f787088d5	Validate culling interval.	9 years ago
Kevin Bates	4000838321	Cull idle kernels after specified period	9 years ago
Min RK	2a5f8d2e17	add token auth to kernel websocket test	9 years ago
Min RK	75a0d71c28	give the server a second to notice that the websocket closed ... because travis is slow	9 years ago
Min RK	8b76de80be	move tz utils to top-level private module ._tz	9 years ago
Min RK	9f55a857f3	use Z isoformat in UTC timestamps ... instead of +00:00	9 years ago
Min RK	d45cb2ad80	track active kernel connections	9 years ago
Min RK	b472c55d9b	add activity watching to kernels ... adds to kernel REST model: - last_activity: datetime-aware timestamp - execution_state: from last recorded status message	9 years ago
Min RK	8e141df890	run Python tests with a token ... exercises token auth	9 years ago
Min RK	fe86b65183	json_errors should be outermost decorator on API handlers ... in order to apply to 403 raised by web.authenticated	9 years ago
Min RK	3be8697b48	iopub rate limit refinements ... - reset counter on status: idle (avoids Run All running into limits) - fix rate measurements to exclude messages not sent due to rate limiting - fix formatting of rate-exceeded messages, tweak text a bit - after pausing output, don't resume sending until rate has dropped 20% below limit, to avoid rapid back & forth across the limit	9 years ago
Sylvain Corlay	a91af4d642	Update traitlets API	10 years ago
Min RK	b6e8a3732e	Allow replacing stale websocket connections ... instead of refusing duplecate connections, since reconnect on flaky network is more likely.	10 years ago
Min RK	4fe10b8142	Prevent session_id collisions ... Keep a registry of open sessions, and refuse to open duplicates.	10 years ago
Min RK	1c47a3dbb5	channel.closed is a method ... since we weren't calling it, the restart channel was never closed, causing process teardown to hang *sometimes*, depending on garbage collection.	10 years ago
Diego Mesa	482ed8a5a3	updating log.warn to log.warning	10 years ago
Min RK	c280b773fb	use session.deserialize to unpack message for rate limiting ... rather than hardcoding json.loads Messages should **never** be deserialized by any means other than the Session API.	10 years ago
Jonathan Frederic	9e2c95dc07	limit_window -> rate_limit_window	10 years ago
Jonathan Frederic	eb30002a82	Use a set for filter exceptions	10 years ago
Jonathan Frederic	7aaccbd545	Don't filter status, comm_info, kernel_info, and execute_request messages	10 years ago
Jonathan Frederic	c47f3d021b	Fix a bug in the code that parses the parent message from the msg_list	10 years ago
Jonathan Frederic	7232a6a99a	Clarification & cleanup	10 years ago