[ https://issues.apache.org/jira/browse/ZOOKEEPER-2894?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Alexander A. Strelets updated ZOOKEEPER-2894: --------------------------------------------- Description: ZooKeeper C Client *+single thread+* build First of all, ZooKeeper C Client design allows calling _zookeeper_close()_ in two ways: a) from a ZooKeeper callback handler (completion or watcher) which in turn is called through _zookeeper_process()_ b) and from other places -- i.e., when the call-stack does not pass through any of zookeeper mechanics prior to enter into mentioned _zookeeper_close()_ The issue described here below is +specific only to the case (b)+. So, it's Ok with the case (a). When _zookeeper_close()_ is called in the (b) way, the following happens: 1. +If there are requests waiting for responses in _zh.sent_requests_ queue+, they all are removed from this queue and each of them is "completed" with personal fake response having status ZCLOSING. Such fake responses are put into _zh.completions_to_process_ queue. It's Ok 2. But then, _zh.completions_to_process_ queue is left unhandled. +Neither completion callbacks are called, nor dynamic memory allocated for fake responses is freed+ 3. Different structures within _zh_ are dismissed and finally _zh_ is freed This is illustrated on the screenshot attached to this ticket: you may see that the next instruction to execute will be _free(zh)_ while _zh.completions_to_process_ queue is not empty (see the "Variables" tab to the right). Alternatively, the same situation but in the case (a) is handled properly -- i.e., all completion callback handlers are truly called with ZCLOSING and the memory is freed, both for subcases (a.1) when there is a failure like connection-timeout, connection-closed, etc., or (a.2) there is not failure. The reason is that any callback handler (completion or watcher) in the case (a) is called from the _process_completions()_ function which runs in the loop until _zh.completions_to_process_ queue gets empty. So, this function guarantees this queue to be completely processed even if new completions occur during reaction on previously queued completions. *Consequently:* 1. At least there is definitely the +memory leak+ in the case (b) -- all the fake responses put into _zh.completions_to_process_ queue are lost after _free(zh)_ 2. And it looks like a great misbehavior not to call completions on sent requests in the case (b) while they are called with ZCLOSING in the case (a) -- so, I think it's not "by design" but a +completions leak+ +To reproduce the case (b) do the following:+ - open ZooKeeper session, connect to a server, receive and process connected-watch, etc. - then somewhere +from the main events loop+ call for example _zoo_acreate()_ with valid arguments -- it shall return ZOK - then, +immediately after it returned+, call _zookeeper_close()_ - note that completion callback handler for _zoo_acreate()_ will not be called +To reproduce the case (a) do the following:+ - the same as above, open ZooKeeper session, connect to a server, receive and process connected-watch, etc. - the same as above, somewhere from the main events loop call for example _zoo_acreate()_ with valid arguments -- it shall return ZOK - but now don't call _zookeeper_close()_ immediately -- wait for completion callback on the commenced request - when _zoo_acreate()_ completes, +from within its completion callback handler+, call another _zoo_acreate()_ and immediately after it returned call _zookeeper_close()_ - note that completion callback handler for the second _zoo_acreate()_ will be called with ZCLOSING, unlike the case (b) described above To fix this I propose calling to _process_completions()_ from _destroy(zhandle_t *zh)_ as it is done in _handle_error(zhandle_t *zh,int rc)_. was: ZooKeeper C Client *+single thread+* build First of all, ZooKeeper C Client design allows calling _zookeeper_close()_ in two ways: a) from a ZooKeeper callback handler (completion or watcher) which in turn is called through _zookeeper_process()_ b) and from other places -- i.e., when the call-stack does not pass through any of zookeeper mechanics prior to enter into mentioned _zookeeper_close()_ The issue described here below is +specific only to the case (b)+. So, it's Ok with the case (a). When _zookeeper_close()_ is called in the (b) way, the following happens: 1. +If there are requests waiting for responses in _zh.sent_requests_ queue+, they all are removed from this queue and each of them is "completed" with personal fake response having status ZCLOSING. Such fake responses are put into _zh.completions_to_process_ queue. It's Ok 2. But then, _zh.completions_to_process_ queue is left unhandled. +Neither completion callbacks are called, nor dynamic memory allocated for fake responses is freed+ 3. Different structures within _zh_ are dismissed and finally _zh_ is freed This is illustrated on the screenshot attached to this ticket: you may see that the next instruction to execute will be _free(zh)_ while _zh.completions_to_process_ queue is not empty (see the "Variables" tab to the right). Alternatively, the same situation but in the case (a) is handled properly -- i.e., all completion callback handlers are truly called with ZCLOSING and the memory is freed, both for subcases (a.1) when there is a failure like connection-timeout, connection-closed, etc., or (a.2) there is not failure. The reason is that any callback handler (completion or watcher) in the case (a) is called from the _process_completions()_ function which runs in the loop until _zh.completions_to_process_ queue gets empty. So, this function guarantees this queue to be completely processed even if new completions occur during reaction on previously queued completions. *Consequently:* 1. At least there is definitely the +memory leak+ in the case (b) -- all the fake responses put into _zh.completions_to_process_ queue are lost after _free(zh)_ 2. And it looks like a great misbehavior not to call completions on sent requests in the case (b) while they are called with ZCLOSING in the case (a) -- so, I think it's not "by design" but a +completions leak+ +To reproduce the case (b) do the following:+ - open ZooKeeper session, connect to a server, receive and process connected-watch, etc. - then somewhere from the main events loop call for example _zoo_acreate()_ with valid arguments -- it shall return ZOK - then, immediately after it returned, call _zookeeper_close()_ - note that completion callback handler for _zoo_acreate()_ will not be called +To reproduce the case (a) do the following:+ - the same as above, open ZooKeeper session, connect to a server, receive and process connected-watch, etc. - the same as above, somewhere from the main events loop call for example _zoo_acreate()_ with valid arguments -- it shall return ZOK - but now don't call _zookeeper_close()_ immediately -- wait for completion callback on the commenced request - when _zoo_acreate()_ completes, +from within its completion callback handler+, call another _zoo_acreate()_ and immediately after it returned call _zookeeper_close()_ - note that completion callback handler for the second _zoo_acreate()_ will be called with ZCLOSING, unlike the case (b) described above To fix this I propose calling to _process_completions()_ from _destroy(zhandle_t *zh)_ as it is done in _handle_error(zhandle_t *zh,int rc)_. > Memory and completions leak on zookeeper_close. > ----------------------------------------------- > > Key: ZOOKEEPER-2894 > URL: https://issues.apache.org/jira/browse/ZOOKEEPER-2894 > Project: ZooKeeper > Issue Type: Bug > Components: c client > Affects Versions: 3.4.10 > Environment: Linux ubuntu 4.4.0-87-generic > gcc (Ubuntu 5.4.0-6ubuntu1~16.04.4) 5.4.0 20160609 > https://github.com/apache/zookeeper.git > branch-3.4 > Reporter: Alexander A. Strelets > Priority: Critical > Fix For: 3.4.10 > > > ZooKeeper C Client *+single thread+* build > First of all, ZooKeeper C Client design allows calling _zookeeper_close()_ in > two ways: > a) from a ZooKeeper callback handler (completion or watcher) which in turn is > called through _zookeeper_process()_ > b) and from other places -- i.e., when the call-stack does not pass through > any of zookeeper mechanics prior to enter into mentioned _zookeeper_close()_ > The issue described here below is +specific only to the case (b)+. So, it's > Ok with the case (a). > When _zookeeper_close()_ is called in the (b) way, the following happens: > 1. +If there are requests waiting for responses in _zh.sent_requests_ queue+, > they all are removed from this queue and each of them is "completed" with > personal fake response having status ZCLOSING. Such fake responses are put > into _zh.completions_to_process_ queue. It's Ok > 2. But then, _zh.completions_to_process_ queue is left unhandled. +Neither > completion callbacks are called, nor dynamic memory allocated for fake > responses is freed+ > 3. Different structures within _zh_ are dismissed and finally _zh_ is freed > This is illustrated on the screenshot attached to this ticket: you may see > that the next instruction to execute will be _free(zh)_ while > _zh.completions_to_process_ queue is not empty (see the "Variables" tab to > the right). > Alternatively, the same situation but in the case (a) is handled properly -- > i.e., all completion callback handlers are truly called with ZCLOSING and the > memory is freed, both for subcases (a.1) when there is a failure like > connection-timeout, connection-closed, etc., or (a.2) there is not failure. > The reason is that any callback handler (completion or watcher) in the case > (a) is called from the _process_completions()_ function which runs in the > loop until _zh.completions_to_process_ queue gets empty. So, this function > guarantees this queue to be completely processed even if new completions > occur during reaction on previously queued completions. > *Consequently:* > 1. At least there is definitely the +memory leak+ in the case (b) -- all the > fake responses put into _zh.completions_to_process_ queue are lost after > _free(zh)_ > 2. And it looks like a great misbehavior not to call completions on sent > requests in the case (b) while they are called with ZCLOSING in the case (a) > -- so, I think it's not "by design" but a +completions leak+ > +To reproduce the case (b) do the following:+ > - open ZooKeeper session, connect to a server, receive and process > connected-watch, etc. > - then somewhere +from the main events loop+ call for example _zoo_acreate()_ > with valid arguments -- it shall return ZOK > - then, +immediately after it returned+, call _zookeeper_close()_ > - note that completion callback handler for _zoo_acreate()_ will not be called > +To reproduce the case (a) do the following:+ > - the same as above, open ZooKeeper session, connect to a server, receive and > process connected-watch, etc. > - the same as above, somewhere from the main events loop call for example > _zoo_acreate()_ with valid arguments -- it shall return ZOK > - but now don't call _zookeeper_close()_ immediately -- wait for completion > callback on the commenced request > - when _zoo_acreate()_ completes, +from within its completion callback > handler+, call another _zoo_acreate()_ and immediately after it returned call > _zookeeper_close()_ > - note that completion callback handler for the second _zoo_acreate()_ will > be called with ZCLOSING, unlike the case (b) described above > To fix this I propose calling to _process_completions()_ from > _destroy(zhandle_t *zh)_ as it is done in _handle_error(zhandle_t *zh,int > rc)_. -- This message was sent by Atlassian JIRA (v6.4.14#64029)