Mailing List Archive

On Mon, 23 Jun 2008 22:44:22 EDT, Christian Blair said:

> We don't need everyone's traceroute.
> Hardly useful for others anyway unless they're your roommate.

Actually, you have that backwards.

If somebody posts a traceroute from a vastly different network location,
that tells you something. In the thread that you're apparently commenting
on, Jay Hennigan posted a traceroute that *worked* when going via Level3,
which tells us that lightlink.com is in fact up-and-running, and the routing
issue is probably Cogent-specific.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 226 bytes
Desc: not available
URL: <https://puck.nether.net/pipermail/outages/attachments/20080624/949d25ea/attachment.bin>

On Jun 24, 2008, at 1:15 PM, Valdis.Kletnieks at vt.edu wrote:

> On Mon, 23 Jun 2008 22:44:22 EDT, Christian Blair said:
>
>> We don't need everyone's traceroute.
>> Hardly useful for others anyway unless they're your roommate.
>
> Actually, you have that backwards.
>
> If somebody posts a traceroute from a vastly different network
> location,
> that tells you something.


I think the question really is - is this list designed for *reporting*
and confirming outages? Or is it designed for *diagnosing*/discussing/
troubleshooting outages?

If someone reports an outage, there is a degree of working with the
reporter to confirm the outage that is necessary. (i.e. "AIM isn't
working for me" and others either confirming or saying that it does
work for them.) There have, though, been a couple of threads where
this exchange has seemed to go on for a bit more than simple
confirmation and dived more into troubleshooting.

The "reporting/confirming" type of list is a lower-traffic kind of
list that's more announcement/alert-oriented. The "diagnosing/
discussing" type of list could be a much higher-traffic list. Both
are worthwhile types of lists to have... but they also may attract
different subscribers.

Based on a couple of comments I've seen here, I think some folks may
have signed up thinking they were getting the first type of list and
are less interested in the second type of list. They may be more
interested in "alerts" and not interested in "discussion".

I don't have a great solution to offer... I think the "community" here
is probably too new to split into separate "alerts" and "discussion"
lists. I also personally don't care which type of list it is. But I
would suggest this difference in expectations is perhaps at the route
of the original comment and others that have been on the list.

My 2 cents,
Dan

--
Dan York, CISSP, Director of Emerging Communication Technology
Office of the CTO Voxeo Corporation dyork at voxeo.com
Phone: +1-407-455-5859 Skype: danyork http://www.voxeo.com
Blogs: http://blogs.voxeo.com http://www.disruptivetelephony.com

Build voice applications based on open standards.
Find out how at http://www.voxeo.com/free

> -----Original Message-----
> From: outages-bounces at isotf.org [mailto:outages-bounces at isotf.org] On Behalf
> Of Dan York
> Sent: Tuesday, June 24, 2008 10:33 AM
> To: Valdis.Kletnieks at vt.edu
> Cc: outages at isotf.org
> Subject: Re: [outages] List comment
>
>
> On Jun 24, 2008, at 1:15 PM, Valdis.Kletnieks at vt.edu wrote:
>
> > On Mon, 23 Jun 2008 22:44:22 EDT, Christian Blair said:
> >
> >> We don't need everyone's traceroute.
> >> Hardly useful for others anyway unless they're your roommate.
> >
> > Actually, you have that backwards.
> >
> > If somebody posts a traceroute from a vastly different network
> > location,
> > that tells you something.
>
>
> I think the question really is - is this list designed for *reporting*
> and confirming outages? Or is it designed for *diagnosing*/discussing/
> troubleshooting outages?
>
> If someone reports an outage, there is a degree of working with the
> reporter to confirm the outage that is necessary. (i.e. "AIM isn't
> working for me" and others either confirming or saying that it does
> work for them.) There have, though, been a couple of threads where
> this exchange has seemed to go on for a bit more than simple
> confirmation and dived more into troubleshooting.
>
> The "reporting/confirming" type of list is a lower-traffic kind of
> list that's more announcement/alert-oriented. The "diagnosing/
> discussing" type of list could be a much higher-traffic list. Both
> are worthwhile types of lists to have... but they also may attract
> different subscribers.
>
> Based on a couple of comments I've seen here, I think some folks may
> have signed up thinking they were getting the first type of list and
> are less interested in the second type of list. They may be more
> interested in "alerts" and not interested in "discussion".
>
> I don't have a great solution to offer... I think the "community" here
> is probably too new to split into separate "alerts" and "discussion"
> lists. I also personally don't care which type of list it is. But I
> would suggest this difference in expectations is perhaps at the route
> of the original comment and others that have been on the list.
>
> My 2 cents,
> Dan
>

I like your breakdown. How about a general consensus to preface subject lines with either "Outage" or "Discussion" until there is enough momentum to break the list into two?

Regards,

Mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 475 bytes
Desc: not available
URL: <https://puck.nether.net/pipermail/outages/attachments/20080624/65a5adf7/attachment.bin>

Michael K. Smith - Adhost wrote:


> I like your breakdown. How about a general consensus to preface
> subject lines with either "Outage" or "Discussion" until there is
> enough momentum to break the list into two?
I've pretty much be lurker (not being currently employed, and all) but
it appears to me that this list, like others that generate a fair amount
of meta-discussion, has two main constituencies:

People that write about that which interests me.
People that write about that which is a complete and utter waste of packets.

--
Requiescas in pace o email Two identifying characteristics
of System Administrators:
Ex turpi causa non oritur actio Infallibility, and the ability to
learn from their mistakes.
Eppure si rinfresca

ICBM Targeting Information: http://tinyurl.com/4sqczs

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dan York wrote:
>
> On Jun 24, 2008, at 1:15 PM, Valdis.Kletnieks at vt.edu wrote:
>
>> On Mon, 23 Jun 2008 22:44:22 EDT, Christian Blair said:
>>
>>> We don't need everyone's traceroute.
>>> Hardly useful for others anyway unless they're your roommate.
>>
>> Actually, you have that backwards.
>>
>> If somebody posts a traceroute from a vastly different network location,
>> that tells you something.
>
>
> I think the question really is - is this list designed for *reporting*
> and confirming outages? Or is it designed for
> *diagnosing*/discussing/troubleshooting outages?
>
> If someone reports an outage, there is a degree of working with the
> reporter to confirm the outage that is necessary. (i.e. "AIM isn't
> working for me" and others either confirming or saying that it does work
> for them.) There have, though, been a couple of threads where this
> exchange has seemed to go on for a bit more than simple confirmation and
> dived more into troubleshooting.
>
> The "reporting/confirming" type of list is a lower-traffic kind of list
> that's more announcement/alert-oriented. The "diagnosing/discussing"
> type of list could be a much higher-traffic list. Both are worthwhile
> types of lists to have... but they also may attract different subscribers.
>
> Based on a couple of comments I've seen here, I think some folks may
> have signed up thinking they were getting the first type of list and are
> less interested in the second type of list. They may be more interested
> in "alerts" and not interested in "discussion".
- ---------------------------
Great feedback.

Lot of times the outages that are being reported could be specific to
one's local PE which may or may not impact others. In others words, the
outages that are being reported may not be confirmed until we have some
sort of feedback from user community confirming the same which I agree
could lead into outages discussion.
Until we establish that relationship/openness between network operator
community and providers we may have to rely on operator community
feedback confirming outage(s) and at the same time keeping it to a minimum.

Now if providers could be little more forthcoming and start reporting
their maintenance and outages then sure I can see those alerts being
filtered into something like 'outages-alerts' but until that happens I
think we may need to feel the list further.


Thoughts?


regards,
/virendra


>
> I don't have a great solution to offer... I think the "community" here
> is probably too new to split into separate "alerts" and "discussion"
> lists. I also personally don't care which type of list it is. But I
> would suggest this difference in expectations is perhaps at the route of
> the original comment and others that have been on the list.
>
> My 2 cents,
> Dan
>
> --Dan York, CISSP, Director of Emerging Communication Technology
> Office of the CTO Voxeo Corporation dyork at voxeo.com
> Phone: +1-407-455-5859 Skype: danyork http://www.voxeo.com
> Blogs: http://blogs.voxeo.com http://www.disruptivetelephony.com
>
> Build voice applications based on open standards.
> Find out how at http://www.voxeo.com/free
>
>
>
>
>
> _______________________________________________
> outages mailing list
> outages at isotf.org
> https://puck.nether.net/mailman/listinfo/outages
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIYTwUpbZvCIJx1bcRAvlPAJ9IORHlF/ABtRTLntAn/bDCqMGIKQCg6qZY
pEPL6TxqEPBLbNkF8Y6lLTI=
=W5re
-----END PGP SIGNATURE-----

On Tue, Jun 24, 2008 at 10:48:38AM -0700, Michael K. Smith - Adhost wrote:
> I like your breakdown. How about a general consensus to preface
> subject lines with either "Outage" or "Discussion" until there is
> enough momentum to break the list into two?

That trick *never* works.

People who are not interested in discussion -- and those who are, say,
forwarding to a Blackberry -- should probably filter out messages whose
subject lines include "[Rr][Ee]: ".

Cheers,
-- jra
--
Jay R. Ashworth Baylink jra at baylink.com
Designer The Things I Think RFC 2100
Ashworth & Associates http://baylink.pitas.com '87 e24
St Petersburg FL USA http://photo.imageinc.us +1 727 647 1274

Those who cast the vote decide nothing.
Those who count the vote decide everything.
-- (Joseph Stalin)

> I think the question really is - is this list designed for *reporting*
> and confirming outages? Or is it designed for *diagnosing*/discussing/
> troubleshooting outages?

I understood the former. It's called outages I want my outages
pure and simple and not cut with possible outages "is teh interwebs
broken, btw I can't print either" :)

There's enough noise and html already I was considering giving up
but the move to puck gave me hope it'll evolve into something useful
to me. If it doesn't then no problem it's probably useful to someone
else.

regards
brandon

---------- dyork at lodestar2.com wrote: --------------
>> We don't need everyone's traceroute.
>> Hardly useful for others anyway unless they're your roommate.

I don't have a great solution to offer... I think the "community" here
is probably too new to split into separate "alerts" and "discussion"
lists. I also personally don't care which type of list it is. But I
would suggest this difference in expectations is perhaps at the route
of the original comment and others that have been on the list.
---------------------------------------------------------


When a subject becomes uninteresting to a person, all he/she has to do is select based on the subject line and mass delete the thread. It's not too hard. I don't think the list has enough momentum to fragment it already.

It would be nice, though, if folks could clean up the traceroutes a bit before posting. For example, remove the delay times when just showing there's connectivity to a particular destination. Many traceroutes are a hard to read mess.

scott





















































-----------------------------

On 24 Jun 2008, at 14:35, Brandon Butterworth wrote:

>> I think the question really is - is this list designed for
>> *reporting*
>> and confirming outages? Or is it designed for *diagnosing*/
>> discussing/
>> troubleshooting outages?
>
> I understood the former. It's called outages I want my outages
> pure and simple and not cut with possible outages "is teh interwebs
> broken, btw I can't print either" :)

Whilst I appreciate that it's ironic that I'm adding to the noise I am
about to say I would prefer not to see, I found this list most
attractive and useful so far when it was nothing more than brief
statements of things that were broken, with no analysis. Ideally (in
my opinion) any analysis, troubleshooting, drawing of conclusions or
other meta-chat would happen somewhere else where I couldn't see it.


Joe

------- virendra.rode at gmail.com wrote: ------------
Now if providers could be little more forthcoming and start reporting
their maintenance and outages then sure I can see those alerts being
filtered into something like 'outages-alerts' but until that happens I
think we may need to feel the list further.
---------------------------------------------------


I wouldn't hold my breath for that. You might turn awfully blue... ;-)

scott





































-----------------------------

PROPOSAL:

OUTAGE ALERT to all list members
Include link to Wiki of someone wants to dig deeper



Aaron D. Osgood

Streamline Solutions L.L.C

P.O. Box 6115
Falmouth, ME 04105

TEL: 207-781-5561
FAX: 207-781-8067
MOBILE: 207-831-5829
PAGE: 2078315829 at vtext.com
AOLIM: OzCom1
ICQ: 206889374

AOsgood at Streamline-Solutions.net
Blog: http://streamlinesolutionsllc.blogspot.com/
http://www.streamline-solutions.net
http://www.WMDaWARe.com

Introducing Efficiency to Business since 1986.


-----Original Message-----
From: outages-bounces@isotf.org [mailto:outages-bounces@isotf.org] On Behalf
Of Joe Abley
Sent: Tuesday, June 24, 2008 2:40 PM
To: Brandon Butterworth
Cc: outages at isotf.org
Subject: Re: [outages] List comment


On 24 Jun 2008, at 14:35, Brandon Butterworth wrote:

>> I think the question really is - is this list designed for
>> *reporting*
>> and confirming outages? Or is it designed for *diagnosing*/
>> discussing/
>> troubleshooting outages?
>
> I understood the former. It's called outages I want my outages
> pure and simple and not cut with possible outages "is teh interwebs
> broken, btw I can't print either" :)

Whilst I appreciate that it's ironic that I'm adding to the noise I am
about to say I would prefer not to see, I found this list most
attractive and useful so far when it was nothing more than brief
statements of things that were broken, with no analysis. Ideally (in
my opinion) any analysis, troubleshooting, drawing of conclusions or
other meta-chat would happen somewhere else where I couldn't see it.


Joe

_______________________________________________
outages mailing list
outages at isotf.org
https://puck.nether.net/mailman/listinfo/outages

> > I think the question really is - is this list designed for *reporting*
> > and confirming outages? Or is it designed for *diagnosing*/discussing/
> > troubleshooting outages?
>

I like the discussion of any outage related issue, both reporting,
confirming, and troubleshooting.
Now the divulging into areas such as "who still uses ICQ/AIM" and why
they should/should not/whether or not it's secure is way off base.


--

WKH

When I signed up for outages I was really hoping for something like what Sean Donelan used to do for Nanog. He had an occupation where he knew about every major cable cut or issue and reported it to the Nanog mailing list. I really miss that level of outage knowledge that we used to see.

I think that something like outages alert could work even with the current group. The issue is one of is it a real issue that affects a large enough group. For this I think the mailing list with the website could put in a voting idea. Using a footer in the message people could click that to vote and say this is a confirmed issue and select if it's minor or major and regional or larger impact. After a set number of votes hit then the system could send an email to outage-alerts which would be a much lower traffic but a much higher confidence list. That would allow outages to still discuss troubleshooting without imapct the blackberry and pager audience.

I have always looked for a way to be alerted to outages but only one's that truly are an issue. Maybe this is one way to accomplish that ?

Derrick

________________________________

From: outages-bounces@isotf.org on behalf of Jay R. Ashworth
Sent: Tue 6/24/2008 2:29 PM
To: outages at isotf.org
Subject: Re: [outages] List comment



On Tue, Jun 24, 2008 at 10:48:38AM -0700, Michael K. Smith - Adhost wrote:
> I like your breakdown. How about a general consensus to preface
> subject lines with either "Outage" or "Discussion" until there is
> enough momentum to break the list into two?

That trick *never* works.

People who are not interested in discussion -- and those who are, say,
forwarding to a Blackberry -- should probably filter out messages whose
subject lines include "[Rr][Ee]: ".

Cheers,
-- jra
--
Jay R. Ashworth Baylink jra at baylink.com
Designer The Things I Think RFC 2100
Ashworth & Associates http://baylink.pitas.com <http://baylink.pitas.com/> '87 e24
St Petersburg FL USA http://photo.imageinc.us <http://photo.imageinc.us/> +1 727 647 1274

Those who cast the vote decide nothing.
Those who count the vote decide everything.
-- (Joseph Stalin)
_______________________________________________
outages mailing list
outages at isotf.org
https://puck.nether.net/mailman/listinfo/outages


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/outages/attachments/20080624/36687174/attachment.html>

Howdy folks. I don't post much here ,but since this is about the
list, here's my two cents.

I think there are more than two (report, discussion) topics reltated
to an outage. Here's my list, including what I would find informative
(useful) vs. interesting (take it or leave it) vs. annoying (leave
it). This is assuming that the outage is a major network outage
(affects more than a handful of sites). I'm not really interested to
know that myspace or facebook is down again?i get enough of that on
the nanog list.

1) planned outages. These would be informative if they met the
criteria but the fact that AboveNet is doing yet another switch
upgrade at boone doesn't really qualify. Also any planned maintenance
is nto a planned outage.
2) Current outages (reporting outages as they are happening-e.g. So-
and-so is down here-or-there): informative.
3) Troubleshooting. Take it elsewhere. If you want a traceroute from
Burkina Fasso, fine a looking glass. On a sidenote, I would be OK
with someone actually currently working for an outage asking for some
type of help from the list (e.g. a traceroute from specific areas) but
replies should not go to the list
4) Ongoing outage discussion. This can be very informative and
interesting if there's real data on a real outage (e.g. ordered new
parts for X equipment, expect delivery in Y hours, back up in Z
time). Better yet, however, the original outage post should have a
link where updates are provided for those interested.
5) post-mortem. This is always helpful. At best it helps us learn
from our mistakes, at worst it makes me feel better knowing I'm not
the only idiot out here.
6) ad-nauseum post-mortem discussion. This can be between interesting
and annoying?but if kept within a thread I can delete the whole
thread. Mostly these start out interesting and then turn into
opinionated banter.

However many of these there are, I would still like to see only one
thread per outage. If I want to drill further, I can read all the
replies.

Bottom line: I read this list before I read the news, so things like
the issues with florida a few months back, or a network outage, etc,
are what I would like to see here. After the outage, a description
of what happened is also nice to have, as is ongoing reports for
_major_ outages. Anything more becomes noise.

Best,
Y

--
Mickey Panayiotakis - mickey at srtdata.com
http://www.grassroots.org/
P Think before you print.




On Jun 24, 2008, at 3:28 PM, Derrick Bennett wrote:

> When I signed up for outages I was really hoping for something like
> what Sean Donelan used to do for Nanog. He had an occupation where
> he knew about every major cable cut or issue and reported it to the
> Nanog mailing list. I really miss that level of outage knowledge
> that we used to see.
>
> I think that something like outages alert could work even with the
> current group. The issue is one of is it a real issue that affects a
> large enough group. For this I think the mailing list with the
> website could put in a voting idea. Using a footer in the message
> people could click that to vote and say this is a confirmed issue
> and select if it's minor or major and regional or larger impact.
> After a set number of votes hit then the system could send an email
> to outage-alerts which would be a much lower traffic but a much
> higher confidence list. That would allow outages to still discuss
> troubleshooting without imapct the blackberry and pager audience.
>
> I have always looked for a way to be alerted to outages but only
> one's that truly are an issue. Maybe this is one way to accomplish
> that ?
>
> Derrick
>
> From: outages-bounces at isotf.org on behalf of Jay R. Ashworth
> Sent: Tue 6/24/2008 2:29 PM
> To: outages at isotf.org
> Subject: Re: [outages] List comment
>
> On Tue, Jun 24, 2008 at 10:48:38AM -0700, Michael K. Smith - Adhost
> wrote:
> > I like your breakdown. How about a general consensus to preface
> > subject lines with either "Outage" or "Discussion" until there is
> > enough momentum to break the list into two?
>
> That trick *never* works.
>
> People who are not interested in discussion -- and those who are, say,
> forwarding to a Blackberry -- should probably filter out messages
> whose
> subject lines include "[Rr][Ee]: ".
>
> Cheers,
> -- jra
> --
> Jay R. Ashworth Baylink jra at baylink.com
> Designer The Things I
> Think RFC 2100
> Ashworth & Associates http://
> baylink.pitas.com '87 e24
> St Petersburg FL USA http://photo.imageinc.us +1
> 727 647 1274
>
> Those who cast the vote decide nothing.
> Those who count the vote decide everything.
> -- (Joseph Stalin)
> _______________________________________________
> outages mailing list
> outages at isotf.org
> https://puck.nether.net/mailman/listinfo/outages
>
> _______________________________________________
> outages mailing list
> outages at isotf.org
> https://puck.nether.net/mailman/listinfo/outages

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/outages/attachments/20080624/505991fa/attachment-0001.html>

On Tue, Jun 24, 2008 at 02:40:21PM -0400, Joe Abley wrote:
> Whilst I appreciate that it's ironic that I'm adding to the noise I am
> about to say I would prefer not to see, I found this list most
> attractive and useful so far when it was nothing more than brief
> statements of things that were broken, with no analysis. Ideally (in
> my opinion) any analysis, troubleshooting, drawing of conclusions or
> other meta-chat would happen somewhere else where I couldn't see it.

I agree. The only time I need this list is when I am out working
a live webcast for JP Morgan as part of a contract crew and I don't
have time to read email. I typically only check outages during my
12-14 hour shift so that I can tell my boss why some CEO can't see
the presentation his CFO just made.

I want the facts, naked and pure and above all, short.

On Tue, Jun 24, 2008 at 11:25:24AM -0700, virendra rode // wrote:
> Now if providers could be little more forthcoming and start reporting
> their maintenance and outages then sure I can see those alerts being
> filtered into something like 'outages-alerts' but until that happens I
> think we may need to feel the list further.

That trick never works *either*.

I suspect what we're going to have to do is set up an outage-notify list,
with a companion -post (as nanog does), and let people volunteer to
summarize from the discussion list to -notify once some actual facts
have been gleaned from the discussions.

So, for last week, we might have seen

Service: AIM
Problem: Clients won't connect
Scope: Nationwide
Since: 1015
Reporter: jra at baylink.com
Confirmed: 12 users
Official? No

Service: AIM
Problem: Netsplit
Scope: Certain servers; correlation unclear
Since: 1015
Reporter: up at 3.am
Confirmed: 6 users
Official? No

Service: L3 transport USA:NE
Problem: Routing loop; link backhoed
Scope: AS 18467
Since: 1330
Reporter: somebody at somewhere.com
Confirmed: L3 ticket 772893-338
Official? Yes

each in an email from whomever did the summary.

Would this serve?

Cheers,
-- jra
--
Jay R. Ashworth Baylink jra at baylink.com
Designer The Things I Think RFC 2100
Ashworth & Associates http://baylink.pitas.com '87 e24
St Petersburg FL USA http://photo.imageinc.us +1 727 647 1274

Those who cast the vote decide nothing.
Those who count the vote decide everything.
-- (Joseph Stalin)

Unless we have our own NOC or staff it with volunteers like SANS, it's going
to be hard to make sure that such notifications are timely.

Unfortunately, e-mail is where it's at, and I think adding and
outages-discussion listserv would solve about 75% of the noise concern for
those who want to know *just* about outages.

Frank

-----Original Message-----
From: outages-bounces@isotf.org [mailto:outages-bounces@isotf.org] On Behalf
Of Jay R. Ashworth
Sent: Tuesday, June 24, 2008 3:32 PM
To: outages at isotf.org
Subject: Re: [outages] List comment

On Tue, Jun 24, 2008 at 11:25:24AM -0700, virendra rode // wrote:
> Now if providers could be little more forthcoming and start reporting
> their maintenance and outages then sure I can see those alerts being
> filtered into something like 'outages-alerts' but until that happens I
> think we may need to feel the list further.

That trick never works *either*.

I suspect what we're going to have to do is set up an outage-notify list,
with a companion -post (as nanog does), and let people volunteer to
summarize from the discussion list to -notify once some actual facts
have been gleaned from the discussions.

So, for last week, we might have seen

Service: AIM
Problem: Clients won't connect
Scope: Nationwide
Since: 1015
Reporter: jra at baylink.com
Confirmed: 12 users
Official? No

Service: AIM
Problem: Netsplit
Scope: Certain servers; correlation unclear
Since: 1015
Reporter: up at 3.am
Confirmed: 6 users
Official? No

Service: L3 transport USA:NE
Problem: Routing loop; link backhoed
Scope: AS 18467
Since: 1330
Reporter: somebody at somewhere.com
Confirmed: L3 ticket 772893-338
Official? Yes

each in an email from whomever did the summary.

Would this serve?

Cheers,
-- jra
--
Jay R. Ashworth Baylink
jra at baylink.com
Designer The Things I Think RFC
2100
Ashworth & Associates http://baylink.pitas.com '87
e24
St Petersburg FL USA http://photo.imageinc.us +1 727 647
1274

Those who cast the vote decide nothing.
Those who count the vote decide everything.
-- (Joseph Stalin)
_______________________________________________
outages mailing list
outages at isotf.org
https://puck.nether.net/mailman/listinfo/outages

I like it - be careful, Jay - you are close to making sense

Aaron D. Osgood

Streamline Solutions L.L.C

P.O. Box 6115
Falmouth, ME 04105

TEL: 207-781-5561
FAX: 207-781-8067
MOBILE: 207-831-5829
PAGE: 2078315829 at vtext.com
AOLIM: OzCom1
ICQ: 206889374

AOsgood at Streamline-Solutions.net
Blog: http://streamlinesolutionsllc.blogspot.com/
http://www.streamline-solutions.net
http://www.WMDaWARe.com

Introducing Efficiency to Business since 1986.


-----Original Message-----
From: outages-bounces@isotf.org [mailto:outages-bounces@isotf.org] On Behalf
Of Jay R. Ashworth
Sent: Tuesday, June 24, 2008 4:32 PM
To: outages at isotf.org
Subject: Re: [outages] List comment

On Tue, Jun 24, 2008 at 11:25:24AM -0700, virendra rode // wrote:
> Now if providers could be little more forthcoming and start reporting
> their maintenance and outages then sure I can see those alerts being
> filtered into something like 'outages-alerts' but until that happens I
> think we may need to feel the list further.

That trick never works *either*.

I suspect what we're going to have to do is set up an outage-notify list,
with a companion -post (as nanog does), and let people volunteer to
summarize from the discussion list to -notify once some actual facts
have been gleaned from the discussions.

So, for last week, we might have seen

Service: AIM
Problem: Clients won't connect
Scope: Nationwide
Since: 1015
Reporter: jra at baylink.com
Confirmed: 12 users
Official? No

Service: AIM
Problem: Netsplit
Scope: Certain servers; correlation unclear
Since: 1015
Reporter: up at 3.am
Confirmed: 6 users
Official? No

Service: L3 transport USA:NE
Problem: Routing loop; link backhoed
Scope: AS 18467
Since: 1330
Reporter: somebody at somewhere.com
Confirmed: L3 ticket 772893-338
Official? Yes

each in an email from whomever did the summary.

Would this serve?

Cheers,
-- jra
--
Jay R. Ashworth Baylink
jra at baylink.com
Designer The Things I Think RFC
2100
Ashworth & Associates http://baylink.pitas.com '87
e24
St Petersburg FL USA http://photo.imageinc.us +1 727 647
1274

Those who cast the vote decide nothing.
Those who count the vote decide everything.
-- (Joseph Stalin)
_______________________________________________
outages mailing list
outages at isotf.org
https://puck.nether.net/mailman/listinfo/outages

I don't have a problem with discussion because I use a mail client (mutt)
that nicely threads discussion. So if I see a reported outage (and
subsequent discussion/analysis) that I care about: I read it all. If I
see one that that I don't care about: it's trivially easy to ignore.

So as long as the "Subject" line is apropos (and is modified if it
becomes apparent that the actual outage doesn't fit the orginally
chosen subject) then I'd have no problem keeping up with this list
even if volume increased many times over.

And while the wiki is nice, I think it best to keep outage-related
discussions in a text-only forum that can be read and written by
people on alternative/slow/low-bandwidth connections -- because they
may well be compelled to use those *because* of an outage.

---Rsk

On Tue, Jun 24, 2008 at 02:40:21PM -0400, Joe Abley wrote:
> Whilst I appreciate that it's ironic that I'm adding to the noise I am
> about to say I would prefer not to see, I found this list most
> attractive and useful so far when it was nothing more than brief
> statements of things that were broken, with no analysis. Ideally (in
> my opinion) any analysis, troubleshooting, drawing of conclusions or
> other meta-chat would happen somewhere else where I couldn't see it.

Can we set up -discuss and mung the reply-to header on -notify to point to it?

Cheers,
-- jra
--
Jay R. Ashworth Baylink jra at baylink.com
Designer The Things I Think RFC 2100
Ashworth & Associates http://baylink.pitas.com '87 e24
St Petersburg FL USA http://photo.imageinc.us +1 727 647 1274

Those who cast the vote decide nothing.
Those who count the vote decide everything.
-- (Joseph Stalin)

On Tue, Jun 24, 2008 at 11:39:31AM -0700, Scott Weeks wrote:
> It would be nice, though, if folks could clean up the traceroutes
> a bit before posting. For example, remove the delay times when just
> showing there's connectivity to a particular destination. Many
> traceroutes are a hard to read mess.

Let me suggest 'mtr -r -c 5'.

-c 10, if you have the patience.

Cheers,
-- jra
--
Jay R. Ashworth Baylink jra at baylink.com
Designer The Things I Think RFC 2100
Ashworth & Associates http://baylink.pitas.com '87 e24
St Petersburg FL USA http://photo.imageinc.us +1 727 647 1274

Those who cast the vote decide nothing.
Those who count the vote decide everything.
-- (Joseph Stalin)

Extremely useful - as much as discussion is interesting and informative,
for pager/blackberry/smartphone usefulness, Jay's format is probably
most useable day to day. A lot of work though for someone(s).

Thus, the two list idea -notify and -post or -discussion is very
appealing. Aggregated posts as laid out below could come later (probably
on discussion)

Jay R. Ashworth wrote:

> I suspect what we're going to have to do is set up an outage-notify list,
> with a companion -post (as nanog does), and let people volunteer to
> summarize from the discussion list to -notify once some actual facts
> have been gleaned from the discussions.
>
> So, for last week, we might have seen
>
> Service: AIM
> Problem: Clients won't connect
> Scope: Nationwide
> Since: 1015
> Reporter: jra at baylink.com
> Confirmed: 12 users
> Official? No


--
This message has been scanned by MailScanner

Derrick Bennett wrote:
> When I signed up for outages I was really hoping for something like what
> Sean Donelan used to do for Nanog. He had an occupation where he knew
> about every major cable cut or issue and reported it to the Nanog
> mailing list. I really miss that level of outage knowledge that we used
> to see.
>
> I think that something like outages alert could work even with the
> current group. The issue is one of is it a real issue that affects a
> large enough group. For this I think the mailing list with the website
> could put in a voting idea. Using a footer in the message people could
> click that to vote and say this is a confirmed issue and select if it's
> minor or major and regional or larger impact. After a set number of
> votes hit then the system could send an email to outage-alerts which
> would be a much lower traffic but a much higher confidence list. That
> would allow outages to still discuss troubleshooting without imapct the
> blackberry and pager audience.
>
> I have always looked for a way to be alerted to outages but only one's
> that truly are an issue. Maybe this is one way to accomplish that ?

An outage is an issue if you're on one side of it trying to reach
resources on the other side. If not, then it's not.

I was chastised for posting a traceroute here in an attempt to resolve
the question as to whether a route loop was within Cogent or at the
destination AS. IMNSHO this type of thing helps to define and isolate
the outage.

--
Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net
Impulse Internet Service - http://www.impulse.net/
Your local telephone and internet company - 805 884-6323 - WB6RDV

On Tue, 24 Jun 2008, Dan York wrote:
>
> On Jun 24, 2008, at 1:15 PM, Valdis.Kletnieks at vt.edu wrote:
>
>> On Mon, 23 Jun 2008 22:44:22 EDT, Christian Blair said:
>>
>>> We don't need everyone's traceroute.
>>> Hardly useful for others anyway unless they're your roommate.
>>
>> Actually, you have that backwards.
>>
>> If somebody posts a traceroute from a vastly different network location,
>> that tells you something.
>
>
> I think the question really is - is this list designed for *reporting* and
> confirming outages? Or is it designed for *diagnosing*/discussing/
> troubleshooting outages?
>
> If someone reports an outage, there is a degree of working with the reporter
> to confirm the outage that is necessary. (i.e. "AIM isn't working for me"
> and others either confirming or saying that it does work for them.) There
> have, though, been a couple of threads where this exchange has seemed to go
> on for a bit more than simple confirmation and dived more into
> troubleshooting.
>
> The "reporting/confirming" type of list is a lower-traffic kind of list
> that's more announcement/alert-oriented. The "diagnosing/discussing" type of
> list could be a much higher-traffic list. Both are worthwhile types of
> lists to have... but they also may attract different subscribers.
>
> Based on a couple of comments I've seen here, I think some folks may have
> signed up thinking they were getting the first type of list and are less
> interested in the second type of list. They may be more interested in
> "alerts" and not interested in "discussion".
>
> I don't have a great solution to offer... I think the "community" here is
> probably too new to split into separate "alerts" and "discussion" lists. I
> also personally don't care which type of list it is. But I would suggest
> this difference in expectations is perhaps at the route of the original
> comment and others that have been on the list.
>

I think you covered it best. While outages does have some operational
troubleshooting and peer-support going on, it is under the same threads.

While imperfect, I agree completely that the community is too new to
split. I think peer moderation and mail filters worked pretty good for us
so far.

Let's see if it keeps working, then I guess we can turn to
meta-discussion.


> My 2 cents,
> Dan
>
> --
> Dan York, CISSP, Director of Emerging Communication Technology
> Office of the CTO Voxeo Corporation dyork at voxeo.com
> Phone: +1-407-455-5859 Skype: danyork http://www.voxeo.com
> Blogs: http://blogs.voxeo.com http://www.disruptivetelephony.com
>
> Build voice applications based on open standards.
> Find out how at http://www.voxeo.com/free
>
>
>
>
>
> _______________________________________________
> outages mailing list
> outages at isotf.org
> https://puck.nether.net/mailman/listinfo/outages
>