IRC log of ws-ra on 2010-06-08

Timestamps are in UTC.

19:21:20 [RRSAgent]
RRSAgent has joined #ws-ra
19:21:20 [RRSAgent]
logging to http://www.w3.org/2010/06/08-ws-ra-irc
19:21:22 [trackbot]
RRSAgent, make logs public
19:21:22 [Zakim]
Zakim has joined #ws-ra
19:21:24 [trackbot]
Zakim, this will be WSRA
19:21:24 [Zakim]
ok, trackbot; I see WS_WSRA()3:30PM scheduled to start in 9 minutes
19:21:25 [trackbot]
Meeting: Web Services Resource Access Working Group Teleconference
19:21:25 [trackbot]
Date: 08 June 2010
19:22:20 [DaveS]
DaveS has joined #ws-ra
19:24:55 [dug]
dug has joined #ws-ra
19:27:07 [Zakim]
WS_WSRA()3:30PM has now started
19:27:14 [Zakim]
+Doug_Davis
19:27:43 [Zakim]
+Bob_Freund
19:27:51 [gpilz]
gpilz has joined #ws-ra
19:27:57 [Zakim]
+ +1.831.713.aaaa
19:28:16 [asoldano]
asoldano has joined #ws-ra
19:28:36 [Zakim]
+??P11
19:30:02 [MartinC]
MartinC has joined #ws-ra
19:30:32 [Tom_Rutt]
Tom_Rutt has joined #ws-ra
19:31:07 [Zakim]
+Tom_Rutt
19:31:51 [Zakim]
+Wu_Chou
19:31:51 [W3C]
regrets+ Katy Warr
19:32:13 [Zakim]
+ +1.571.262.aabb
19:32:45 [wuchou]
wuchou has joined #ws-ra
19:33:06 [Zakim]
+ +03531498aacc
19:33:30 [MartinC]
zakim, +0353 is MartinC
19:33:30 [Zakim]
+MartinC; got it
19:33:57 [dug]
echo echo echo
19:34:03 [li]
li has joined #ws-ra
19:34:57 [asoldano]
is the conf code still the same?
19:35:01 [dug]
yes
19:35:07 [dug]
zakim, mute MartinC
19:35:07 [Zakim]
MartinC should now be muted
19:35:15 [dug]
just a guess :-)
19:36:21 [Zakim]
+ +1.408.970.aadd
19:36:44 [Fred_Maciel]
Fred_Maciel has joined #ws-ra
19:36:53 [Zakim]
+ +1.800.456.aaee
19:37:08 [dug]
wow - 800 #
19:37:13 [W3C]
present+ Rick Landau
19:37:24 [Zakim]
+ +039331574aaff
19:38:18 [W3C]
scribe: aoldano
19:38:29 [Zakim]
- +039331574aaff
19:38:33 [dug]
Agenda: http://lists.w3.org/Archives/Public/public-ws-resource-access/2010Jun/0002.html
19:38:42 [asoldano]
my line has fallen,... redialing
19:40:10 [Fred_Maciel]
Does this work?
19:40:27 [W3C]
scribe: Fred Maciel
19:40:55 [Fred_Maciel]
topic: approval of agenda
19:41:03 [Fred_Maciel]
No opposition, agenda approved.
19:41:11 [Fred_Maciel]
topic: approval of the minutes
19:41:18 [Fred_Maciel]
No opposition, minutes approved
19:41:36 [Zakim]
+ +039331574aagg
19:41:58 [Fred_Maciel]
topic: issue 9610
19:41:59 [asoldano]
aagg is asoldano
19:42:04 [Zakim]
+[Oracle]
19:42:08 [asoldano]
Zakim, aagg is asoldano
19:42:08 [Zakim]
+asoldano; got it
19:42:14 [Zakim]
- +1.571.262.aabb
19:42:32 [Zakim]
+ +1.571.262.aahh
19:43:51 [Vikas]
Vikas has joined #ws-ra
19:44:49 [DaveS]
q+
19:45:01 [dug]
q+
19:45:02 [MartinC]
q+
19:45:02 [Fred_Maciel]
Heartbeat is in fact a keepalive.
19:46:19 [Fred_Maciel]
Rate is expected to be long, usually around 5 minutes, so no undue burden on the network.
19:50:45 [gpilz]
q+
19:50:52 [gpilz]
q-
19:51:56 [w3c1]
w3c1 has joined #ws-ra
19:55:41 [wuchou]
q+
19:55:42 [DaveS]
Question: Is the current mechanism only one way, or is the subscriber also tested for being allive by the event deliverer?
19:55:45 [W3C]
ack dave
19:56:58 [MartinC]
where are my skis
19:57:19 [DaveS]
q+
19:57:22 [W3C]
ack dug
20:00:47 [MartinC]
zakim, unmute MartinC
20:00:47 [Zakim]
MartinC should no longer be muted
20:01:01 [W3C]
ack martin
20:01:28 [W3C]
http 1.1 specifies "persistent" connections
20:01:52 [W3C]
http 1.0 had a keep alive mechanism, but it got useless in http 1.1
20:02:14 [W3C]
tcp keep alive is a 60 byte request
20:02:32 [W3C]
but it is widely ignored
20:03:05 [dug]
I would prefer a KeepAlive notification and just make sure your filter includes it
20:03:10 [dug]
zamik, mute Martin
20:03:15 [dug]
zamik, mute MartinC
20:03:25 [dug]
zakim, mute Martin
20:03:25 [Zakim]
MartinC should now be muted
20:03:36 [W3C]
zamik, nobits zaraih!
20:03:47 [dug]
can't type today
20:04:52 [W3C]
tradeoff on persistent connections is the thread needed for each one
20:05:06 [dug]
I can't help but wonder why we're so concerned about a Sub vanishing but not a Transfer resource?
20:05:24 [MartinC]
but this is a fault tolerance issue
20:05:31 [W3C]
ack wu
20:05:59 [dug]
q+
20:06:16 [MartinC]
q+
20:07:10 [W3C]
ack dave
20:09:17 [dug]
me too :-)
20:10:01 [dug]
well, w/o RM the heartbeat could be lost
20:11:23 [MartinC]
this seems to be about connection availability not about server/subscription. we dont explictly tie a subscription to a single connection
20:14:08 [DaveS]
q+
20:14:09 [W3C]
ack dug
20:16:21 [wuchou]
q+
20:20:24 [gpilz]
If the event source terminates a subscription unexpectedly, and it supports the use of the EndTo EPR, and the EndTo EPR was present in the Subscribe message for that subscription (seeĀ 4.1 Subscribe), the SubscriptionEnd message MUST be sent to the endpoint reference indicated by that EPR, if possible.
20:22:56 [MartinC]
zakim, unmute MartinC
20:22:56 [Zakim]
MartinC should no longer be muted
20:24:00 [W3C]
ack martin
20:24:22 [asoldano]
is the status of the communication link really a concern of ws-ra?
20:24:39 [dug]
I didn't think so
20:24:59 [dug]
zakim, mute MartinC
20:24:59 [Zakim]
MartinC should now be muted
20:25:23 [MartinC]
stop muting me;)
20:25:25 [dug]
SubEnd will tell you that
20:25:27 [W3C]
ack dave
20:25:31 [dug]
LOL sorry but the echo is bad
20:26:30 [MartinC]
of course the alive is only sent if the subscription is active, but it only tests the connection
20:26:57 [MartinC]
no alive will be sent if a subscription has expired
20:27:34 [asoldano]
lol
20:27:57 [W3C]
q+
20:28:19 [W3C]
+1 dave
20:28:36 [MartinC]
+1
20:28:51 [W3C]
ack wu
20:30:35 [W3C]
ack w3
20:32:44 [Tom_Rutt]
q+
20:32:54 [MartinC]
go idea bob
20:33:01 [MartinC]
s/go/good/
20:33:57 [Fred_Maciel]
Summary: there are different kinds of deployment scenarios that will need different kinds of periodic messages. Might need to be a special message type dependent on the domain of implementation.
20:36:13 [dug]
only issue with a heartbeat event is that you have no control over the timing of it
20:36:42 [dug]
s/event/event sent due to a filter/
20:40:28 [DaveS]
q+
20:40:49 [W3C]
ack tom
20:40:50 [dug]
bob - didn't you mean all of this under one subscription?
20:41:07 [DaveS]
I did.
20:44:36 [W3C]
ack dave
20:44:56 [wuchou]
q+
20:45:30 [dug]
we'd need to modify the Filter stuff to have more than one filter (and make them an OR), no?
20:47:07 [dug]
we could define some AND and OR filters, then people could mix-n-match to get the exact events they want
20:48:15 [dug]
q+
20:48:25 [W3C]
q+
20:49:08 [W3C]
ack wu
20:49:12 [W3C]
ack dug
20:52:09 [W3C]
ack w3
20:53:52 [MartinC]
i still think the cleanest solution here is to define a special event
20:53:59 [MartinC]
in some other spec!
20:55:12 [MartinC]
there are problems with all proposals ina general context
20:56:54 [Tom_Rutt]
q+
20:57:24 [W3C]
ack tom
20:58:06 [MartinC]
throttling and aliveness are orthogonal concepts
20:59:48 [W3C]
a) heartbeat proposal by gil/wsman
21:00:20 [W3C]
b) domain defined event type that could be subscribed / filter
21:00:51 [W3C]
c) device/domain fixed begavior
21:01:15 [W3C]
d) add interval & count to getstatus/verify
21:01:23 [W3C]
e) cwna
21:01:33 [dug]
f) cwna & kill current verify
21:03:21 [li]
s/begavior/behavior/g
21:03:22 [MartinC]
me have run out of time so can we think about this for next meeting
21:03:27 [Fred_Maciel]
Q: can you live with one of these approaches?
21:03:52 [Fred_Maciel]
RedHat: b c e f
21:04:03 [Fred_Maciel]
IBM: b e f
21:05:00 [Fred_Maciel]
Fujitsu: b d e f
21:05:20 [Fred_Maciel]
Hitachi: abstain
21:05:35 [Fred_Maciel]
Oracle: d e f
21:05:45 [MartinC]
i can live with b, gil couldn't
21:06:03 [Fred_Maciel]
Avaya: a b d e f
21:06:16 [Fred_Maciel]
SoftwareAG: b e f
21:08:34 [MartinC]
next week please!
21:08:55 [dug]
at least B is a more generic solution
21:09:08 [MartinC]
agreed
21:10:11 [Zakim]
-Wu_Chou
21:10:13 [Zakim]
-[Oracle]
21:10:14 [Zakim]
-asoldano
21:10:17 [Zakim]
-Bob_Freund
21:10:18 [Zakim]
-??P11
21:10:19 [Fred_Maciel]
Continue discussion next week on "b", "e" or "f".
21:10:20 [MartinC]
zzzzzzzzzzzzzzz
21:10:22 [Zakim]
- +1.571.262.aahh
21:10:25 [Zakim]
-MartinC
21:10:26 [Zakim]
-Doug_Davis
21:10:30 [Zakim]
- +1.408.970.aadd
21:10:31 [Zakim]
-Tom_Rutt
21:10:33 [Zakim]
- +1.800.456.aaee
21:11:12 [asoldano]
asoldano has left #ws-ra
21:12:14 [W3C]
rrsagent, generate minutes
21:12:14 [RRSAgent]
I have made the request to generate http://www.w3.org/2010/06/08-ws-ra-minutes.html W3C
21:12:36 [W3C]
W3C has left #ws-ra
21:14:36 [w3c1]
w3c1 has left #ws-ra
21:15:33 [Zakim]
disconnecting the lone participant, +1.831.713.aaaa, in WS_WSRA()3:30PM
21:15:35 [Zakim]
WS_WSRA()3:30PM has ended
21:15:39 [Zakim]
Attendees were Doug_Davis, Bob_Freund, +1.831.713.aaaa, Tom_Rutt, Wu_Chou, +1.571.262.aabb, +03531498aacc, MartinC, +1.408.970.aadd, +1.800.456.aaee, +039331574aaff,
21:15:42 [Zakim]
... +039331574aagg, [Oracle], asoldano, +1.571.262.aahh
21:53:47 [gpilz]
gpilz has left #ws-ra
23:10:32 [Zakim]
Zakim has left #ws-ra