[Cake] cake stats 1+ stuff not big on

[Cake] cake stats 1+ stuff not big on

[Dave Taht]

what does 1+ mean?

I do not like this because it really messes with simple parsers of the
resulting data. Key/vaue pairs only please. I have similar feelings
about "way inds", and would prefer that sort of stuff way-inds.

root@lakshmi:~/git/sch_cake# tc -s qdisc show dev wlan0
qdisc cake 8001: root refcnt 5 unlimited squash flows raw
 Sent 53144408 bytes 154976 pkt (dropped 4280, overlimits 0 requeues 467)
 backlog 30554b 0p requeues 467
           Class 0
  rate          0bit
  target       5.0ms
interval     100.0ms
Pk delay        17us # no like this also for parsing reasons.
Av delay         3us
Sp delay         0us
  pkts         56588
way inds        4814
way miss          90
way cols           0
  bytes     14872621
  drops         3417
  marks            0
  flows      1+    2 # no like this for parsing reasons
lastlen           66
max len         1514


-- 
Dave Täht
Do you want faster, better, wifi? https://www.patreon.com/dtaht

Re: [Cake] cake stats 1+ stuff not big on

[Kevin Darbyshire-Bryant]

[-- Attachment #1: Type: text/plain, Size: 1229 bytes --]



On 04/10/15 23:40, Dave Taht wrote:
> what does 1+ mean?
>
> I do not like this because it really messes with simple parsers of the
> resulting data. Key/vaue pairs only please. I have similar feelings
> about "way inds", and would prefer that sort of stuff way-inds.
>
> root@lakshmi:~/git/sch_cake# tc -s qdisc show dev wlan0
> qdisc cake 8001: root refcnt 5 unlimited squash flows raw
>  Sent 53144408 bytes 154976 pkt (dropped 4280, overlimits 0 requeues 467)
>  backlog 30554b 0p requeues 467
>            Class 0
>   rate          0bit
>   target       5.0ms
> interval     100.0ms
> Pk delay        17us # no like this also for parsing reasons.
> Av delay         3us
> Sp delay         0us
>   pkts         56588
> way inds        4814
> way miss          90
> way cols           0
>   bytes     14872621
>   drops         3417
>   marks            0
>   flows      1+    2 # no like this for parsing reasons
> lastlen           66
> max len         1514
>
>

Would https://github.com/dtaht/tc-adv/pull/6 be acceptable to your
sight? :-)

Actually I've just spotted another little tweak to do....there'll be
another commit in a little while.....like busses these commits ;-)




[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4816 bytes --]

Re: [Cake] cake stats 1+ stuff not big on

[Kevin Darbyshire-Bryant]

[-- Attachment #1: Type: text/plain, Size: 2045 bytes --]



On 05/10/15 09:58, Kevin Darbyshire-Bryant wrote:
>
> On 04/10/15 23:40, Dave Taht wrote:
>> what does 1+ mean?
>>
>> I do not like this because it really messes with simple parsers of the
>> resulting data. Key/vaue pairs only please. I have similar feelings
>> about "way inds", and would prefer that sort of stuff way-inds.
>>
>>
>>
> Would https://github.com/dtaht/tc-adv/pull/6 be acceptable to your
> sight? :-)
>
> Actually I've just spotted another little tweak to do....there'll be
> another commit in a little while.....like busses these commits ;-)
And merged with my freshly bestowed super-cow^H^H^Hmerge powers

I didn't think I suffered from OCD but I have to say that 'max-len' not
lining up is bringing it out..hmmmm!

root@Router:~# tc -s qdisc show dev ifb4eth0
qdisc cake 800b: root refcnt 2 bandwidth 39300Kbit diffserv4 flows noatm
overhead 12
 Sent 161649 bytes 721 pkt (dropped 0, overlimits 47 requeues 0)
 backlog 0b 0p requeues 0
             Bin 0       Bin 1       Bin 2       Bin 3 
  rate     39300Kbit   36843Kbit   29475Kbit    9825Kbit
  target       5.0ms       5.0ms       5.0ms       5.0ms
interval     100.0ms     100.0ms     100.0ms     100.0ms
Pk-delay         0us        82us         0us         0us
Av-delay         0us         7us         0us         0us
Sp-delay         0us         1us         0us         0us
  pkts             0         721           0           0
way-inds           0           0           0           0
way-miss           0         104           0           0
way-cols           0           0           0           0
  bytes            0      161649           0           0
  drops            0           0           0           0
  marks            0           0           0           0
Sp-flows           0           0           0           0
Blkflows           0           2           0           0
last-len           0          82           0           0
max-len            0        1514           0           0









[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4816 bytes --]

Re: [Cake] cake stats 1+ stuff not big on

[Dave Taht]

also please, stay clear of sch_cake today. I am about 3/4s done
through fixing checkpatch... it's painful...

On Mon, Oct 5, 2015 at 2:22 AM, Kevin Darbyshire-Bryant
<kevin@darbyshire-bryant.me.uk> wrote:
>
>
> On 05/10/15 09:58, Kevin Darbyshire-Bryant wrote:
>>
>> On 04/10/15 23:40, Dave Taht wrote:
>>> what does 1+ mean?
>>>
>>> I do not like this because it really messes with simple parsers of the
>>> resulting data. Key/vaue pairs only please. I have similar feelings
>>> about "way inds", and would prefer that sort of stuff way-inds.
>>>
>>>
>>>
>> Would https://github.com/dtaht/tc-adv/pull/6 be acceptable to your
>> sight? :-)
>>
>> Actually I've just spotted another little tweak to do....there'll be
>> another commit in a little while.....like busses these commits ;-)
> And merged with my freshly bestowed super-cow^H^H^Hmerge powers
>
> I didn't think I suffered from OCD but I have to say that 'max-len' not
> lining up is bringing it out..hmmmm!
>
> root@Router:~# tc -s qdisc show dev ifb4eth0
> qdisc cake 800b: root refcnt 2 bandwidth 39300Kbit diffserv4 flows noatm
> overhead 12
>  Sent 161649 bytes 721 pkt (dropped 0, overlimits 47 requeues 0)
>  backlog 0b 0p requeues 0
>              Bin 0       Bin 1       Bin 2       Bin 3
>   rate     39300Kbit   36843Kbit   29475Kbit    9825Kbit
>   target       5.0ms       5.0ms       5.0ms       5.0ms
> interval     100.0ms     100.0ms     100.0ms     100.0ms
> Pk-delay         0us        82us         0us         0us
> Av-delay         0us         7us         0us         0us
> Sp-delay         0us         1us         0us         0us
>   pkts             0         721           0           0
> way-inds           0           0           0           0
> way-miss           0         104           0           0
> way-cols           0           0           0           0
>   bytes            0      161649           0           0
>   drops            0           0           0           0
>   marks            0           0           0           0
> Sp-flows           0           0           0           0
> Blkflows           0           2           0           0
> last-len           0          82           0           0
> max-len            0        1514           0           0
>
>
>
>
>
>
>
>
>
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake
>



-- 
Dave Täht
Do you want faster, better, wifi? https://www.patreon.com/dtaht

Re: [Cake] cake stats 1+ stuff not big on

[Toke Høiland-Jørgensen]

Dave Taht <dave.taht@gmail.com> writes:

> Pk delay        17us # no like this also for parsing reasons.

Meh, this is normal, and not hard to deal with in a parser. And since
this is standard for other tc output, it has to be dealt with anyway.

>   flows      1+    2 # no like this for parsing reasons

Never mind parsing reasons; it doesn't make sense to humans either :P


As a general point, I'd argue against making it harder for humans to
read the output in an attempt to make it "easier" to parse. In that
case, do a full machine-readable output (CSV, json, etc); otherwise
we'll end up with ugly output that is still hard to parse...

-Toke

Re: [Cake] cake stats 1+ stuff not big on

[Kevin Darbyshire-Bryant]

[-- Attachment #1: Type: text/plain, Size: 344 bytes --]



On 05/10/15 10:34, Dave Taht wrote:
> also please, stay clear of sch_cake today. I am about 3/4s done
> through fixing checkpatch... it's painful...
Sorry Dave.  Definitely wasn't going to touch sch_cake today!  A simple
'throw-away' change to tc-adv I thought just about permissible.
Kevin has push mode disabled...honest.
>
>



[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4816 bytes --]

Re: [Cake] cake stats 1+ stuff not big on

[Toke Høiland-Jørgensen]

Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk> writes:

> On 05/10/15 10:34, Dave Taht wrote:
>> also please, stay clear of sch_cake today. I am about 3/4s done
>> through fixing checkpatch... it's painful...
> Sorry Dave.  Definitely wasn't going to touch sch_cake today!  A simple
> 'throw-away' change to tc-adv I thought just about permissible.
> Kevin has push mode disabled...honest.

If only there existed some kind of 'system' to keep track of (maybe
'control') the changes (or 'revisions') that people do to the code to
avoid these kinds of conflicts... ;)

-Toke

Re: [Cake] cake stats 1+ stuff not big on

[Sebastian Moeller]

Hi Toke,


On Oct 5, 2015, at 11:55 , Toke Høiland-Jørgensen <toke@toke.dk> wrote:

> Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk> writes:
> 
>> On 05/10/15 10:34, Dave Taht wrote:
>>> also please, stay clear of sch_cake today. I am about 3/4s done
>>> through fixing checkpatch... it's painful...
>> Sorry Dave.  Definitely wasn't going to touch sch_cake today!  A simple
>> 'throw-away' change to tc-adv I thought just about permissible.
>> Kevin has push mode disabled...honest.
> 
> If only there existed some kind of 'system' to keep track of (maybe
> 'control') the changes (or 'revisions') that people do to the code to
> avoid these kinds of conflicts... ;)

	Some times I just don’t git what you are talking about ;)

Best regards
	Sebastian


> 
> -Toke
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake