trafficserver-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From James Peach <>
Subject Re: git commit: TS-2415: use standard continuations to release UrlRewrite objects
Date Wed, 04 Dec 2013 16:11:21 GMT
On Dec 4, 2013, at 7:37 AM, Leif Hedstrom <> wrote:

> On Dec 3, 2013, at 10:30 PM, wrote:
>> Updated Branches:
>> refs/heads/master 2b6a5f6bb -> 0f9305dc6
>> TS-2415: use standard continuations to release UrlRewrite objects
>> UrlRewrite uses a custom continuation to release the old object
>> after it is no longer needed. We can use library code for this
>> instead. Use new_Deleter to free the old UrlRewrite object.
>>  newTable = NEW(new UrlRewrite());
>>  if (newTable->is_valid()) {
>> -    eventProcessor.schedule_in(new UR_FreerContinuation(rewrite_table), URL_REWRITE_TIMEOUT,
>> +    new_Deleter(rewrite_table, URL_REWRITE_TIMEOUT);
>>    Debug("url_rewrite", "remap.config done reloading!");
>>    ink_atomic_swap(&rewrite_table, newTable);
>>  } else {
> Fwiw, this new_Deleter does not have the same semantics as the old code. The old code
would schedule the deletion on the ET_TASK thread pool, whereas new_Deleter runs on any of
the net-threads. This could mean that we’d block a net-thread for some time if the remap
table is sufficiently large.
> I don’t know if it is a real, detrimental problem, but if I recall we moved the remap.config
reloading / management to ET_TASK because it did have a real problem when loading very large

Oh interesting. All the config processor stuff happens on ET_CALL threads too. The remap.config
loading still happens on the ET_TASK.

> Maybe the right thing is to make new_Deleter use ET_TASK?

Yes this sounds like the right thing to do.

> If there are no task threads configured, that falls back to the old behavior?

Or make sure we always have at least 1 task thread?

View raw message