Amos Jeffries writes:
So the Squid logs can contain negative duration under exactly four
conditions:
...
4) NTP service adjusting the kernel clock backwards.
Since you say this is happening noticably often then its #4 and there is
something funky going on with NTP. The others are all at rare an
Amos Jeffries writes:
> On 6/08/2016 1:03 a.m., Henry S. Thompson wrote:
>> Amos wrote:
>>> HST wrote:
>> ...
Amos wrote:
>> ...
> HST wrote:
>> 3) I'm seeing very small numbers (1 in 10) of negative durations,
>> e.g.
...
1000s of times a day? Seems . . . unlikely.
O
On 6/08/2016 1:03 a.m., Henry S. Thompson wrote:
> Amos wrote:
>> HST wrote:
> ...
>>> Amos wrote:
> ...
HST wrote:
> 3) I'm seeing very small numbers (1 in 10) of negative durations,
> e.g.
>
> 1402744620.389 -70 ... TCP_MISS/200 1882 GET http://... DIRECT/...
>
Amos Jeffries writes:
> On 5/08/2016 12:37 a.m., Henry S. Thompson wrote:
>> Thanks for your patience with this, but still not quite getting it.
>>
>> I thought there were two cases:
>>
>> 1) Client drops the connection before the interaction is complete ==
>> log shows HTTP result code 0
Amos wrote:
> HST wrote:
...
>> Amos wrote:
...
>>> HST wrote:
3) I'm seeing very small numbers (1 in 10) of negative durations, e.g.
1402744620.389 -70 ... TCP_MISS/200 1882 GET http://... DIRECT/...
text/xml
How should these be understood?
>>>
>>>
Amos Jeffries writes:
> On 5/08/2016 12:37 a.m., Henry S. Thompson wrote:
>> Thanks for your patience with this, but still not quite getting it.
>>
>> I thought there were two cases:
>>
>> 1) Client drops the connection before the interaction is complete ==
>> log shows HTTP result code 0
On 5/08/2016 12:37 a.m., Henry S. Thompson wrote:
> Thanks for your patience with this, but still not quite getting it.
>
> I thought there were two cases:
>
> 1) Client drops the connection before the interaction is complete ==
> log shows HTTP result code 000, almost always size 0 as wel
Thanks for your patience with this, but still not quite getting it.
I thought there were two cases:
1) Client drops the connection before the interaction is complete ==
log shows HTTP result code 000, almost always size 0 as well;
2) Interaction completed successfully == HTTP result cod
On 4/08/2016 9:52 p.m., Henry S. Thompson wrote:
> Amos Jeffries writes:
>
>> On 4/08/2016 2:36 a.m., Henry S. Thompson wrote:
>>> I'm trying to do some summary statistics based on log files from our
>>> 2.6.STABLE18 setup.
>>
>> Please upgrade. The current 3.5.20 release can do everything that
>>
Amos Jeffries writes:
> On 4/08/2016 2:36 a.m., Henry S. Thompson wrote:
>> I'm trying to do some summary statistics based on log files from our
>> 2.6.STABLE18 setup.
>
> Please upgrade. The current 3.5.20 release can do everything that
> Squid-2.6 could do, and a lot more useful stuff besides.
On 4/08/2016 2:36 a.m., Henry S. Thompson wrote:
> I'm trying to do some summary statistics based on log files from our
> 2.6.STABLE18 setup.
Please upgrade. The current 3.5.20 release can do everything that
Squid-2.6 could do, and a lot more useful stuff besides.
>
> A range of issues with inte
I'm trying to do some summary statistics based on log files from our
2.6.STABLE18 setup.
A range of issues with interpreting 0 and negative values have arisen:
1) Roughly 6 in every 1000 log lines show an HTTP status code of 000,
about 90% with TCP_MISS, the rest mostly TCP_HIT, e.g.
14
12 matches
Mail list logo
