Someone needs to flog the marketing people at Sprint .. http://www.sprintbiz.com/ip/backbone.html "Sprint's Internet backbone includes redundant equipment and diverse routing to ensure the highest reliability. Leveraging Sprint's advanced SONET network, we have specially engineered our network to take full advantage of SONET reliabiliity. In addition, Sprint uses the Cisco Gigabit Switch routers in its backbone, which translates into greater response time and reliability for your users" oops. ========================================================================== Eric Germann CCTec ekgermann@cctec.com Van Wert, OH 45891 http://www.cctec.com Ph: 419 968 2640 ICQ: 41927048 Fax: 419 968 2641 Network Design, Connectivity & System Integration Services A Microsoft Solution Provider
Someone needs to flog the marketing people at Sprint ..
Marketing is marketing. It's hardly operational tho ;-) Ehud p.s. You have "microsoft" and "solutions" together in your .sig. That generally disqualifies you from slapping sprint's hand.
oops.
========================================================================== Eric Germann CCTec ekgermann@cctec.com Van Wert, OH 45891 http://www.cctec.com Ph: 419 968 2640 ICQ: 41927048 Fax: 419 968 2641 Network Design, Connectivity & System Integration Services A Microsoft Solution Provider
Since this is drifting, a bit, and it's got "OT" in the subject line, I'd like to coin a new term for "collapsed SONET rings": 'Two-in-a-trough' Catchy? Piped through the right marketing folks it could even sound like something you'd want. Charles On Tue, 12 Oct 1999, Ehud Gavron wrote:
Marketing is marketing.
It's hardly operational tho ;-)
Ehud p.s. You have "microsoft" and "solutions" together in your .sig. That generally disqualifies you from slapping sprint's hand.
oops.
========================================================================== Eric Germann CCTec ekgermann@cctec.com Van Wert, OH 45891 http://www.cctec.com Ph: 419 968 2640 ICQ: 41927048 Fax: 419 968 2641 Network Design, Connectivity & System Integration Services A Microsoft Solution Provider
Sign up for our new "Eggs-in-a-Basket" service for 2 years and get your outage resolution for FREE! Michael Heller Sr. Systems Engineer Earthweb, Inc. 212.448.4175 mikeh@earthweb.com On Wed, 13 Oct 1999, Charles Sprickman wrote:
Since this is drifting, a bit, and it's got "OT" in the subject line, I'd like to coin a new term for "collapsed SONET rings":
'Two-in-a-trough'
Catchy? Piped through the right marketing folks it could even sound like something you'd want.
Charles
On Tue, 12 Oct 1999, Ehud Gavron wrote:
Marketing is marketing.
It's hardly operational tho ;-)
Ehud p.s. You have "microsoft" and "solutions" together in your .sig. That generally disqualifies you from slapping sprint's hand.
oops.
========================================================================== Eric Germann CCTec ekgermann@cctec.com Van Wert, OH 45891 http://www.cctec.com Ph: 419 968 2640 ICQ: 41927048 Fax: 419 968 2641 Network Design, Connectivity & System Integration Services A Microsoft Solution Provider
On Wed, Oct 13, 1999 at 09:30:38AM -0400, Mike Heller wrote:
Sign up for our new "Eggs-in-a-Basket" service for 2 years and get your outage resolution for FREE!
ITYM "eggs-in-a-casket" semi-operational: in the event that path inside a SONET ring switches from active to protect and back, etc...does this generally show up on a a router as a series of errors, an interface bounce, or not at all? HTH, Sam -- Those who do not understand Unix are condemned to reinvent it, poorly. -- Henry Spencer
I'm guessing that the routers won't notice a cutover, it's too fast. Could someone with more operational experience with SONET confirm this? At 02:27 PM 10/13/99 +0000, Sam Thomas wrote:
On Wed, Oct 13, 1999 at 09:30:38AM -0400, Mike Heller wrote:
Sign up for our new "Eggs-in-a-Basket" service for 2 years and get your outage resolution for FREE!
ITYM "eggs-in-a-casket"
semi-operational: in the event that path inside a SONET ring switches from active to protect and back, etc...does this generally show up on a a router as a series of errors, an interface bounce, or not at all?
HTH, Sam
-- Those who do not understand Unix are condemned to reinvent it, poorly. -- Henry Spencer
Michael Heller Sr. Systems Engineer Earthweb, Inc. 212.448.4175 mikeh@earthweb.com
50ms is fast? You'd drop sevral packets I'd guess. In message <3.0.5.32.19991013115235.00c51100@fox>, Michael Heller writes:
I'm guessing that the routers won't notice a cutover, it's too fast. Could someone with more operational experience with SONET confirm this?
At 02:27 PM 10/13/99 +0000, Sam Thomas wrote:
On Wed, Oct 13, 1999 at 09:30:38AM -0400, Mike Heller wrote:
Sign up for our new "Eggs-in-a-Basket" service for 2 years and get your outage resolution for FREE!
ITYM "eggs-in-a-casket"
semi-operational: in the event that path inside a SONET ring switches from active to protect and back, etc...does this generally show up on a a router as a series of errors, an interface bounce, or not at all?
HTH, Sam
-- Those who do not understand Unix are condemned to reinvent it, poorly. -- Henry Spencer
Michael Heller Sr. Systems Engineer Earthweb, Inc. 212.448.4175 mikeh@earthweb.com
--- jerry@fc.net Freeside/ Insync Internet, Inc.| 512-458-9810 | http://www.fc.net #include <sys/machine/wit/fortune.h>
It shouldn't if you router is configured to not toggle the interface down or you are using the loop-back as update source thing. I don't know for sure if 50ms is enough to cause the interface to detect it being down or not. I suspect it depends a lot on the interface. In message <E11bRM2-000II7-00@rip.psg.com>, Randy Bush writes:
50ms is fast? You'd drop sevral packets I'd guess.
easily survivable. but will bgp flap?
--- jerry@fc.net Freeside/ Insync Internet, Inc.| 512-458-9810 | http://www.fc.net #include <sys/machine/wit/fortune.h>
If you send keepalives every 40 seconds, you will not lose the interface for a 50ms drop. You will incur errors, but you will continue to forward packets. A few will be lost, many will survive. On Wed, Oct 13, 1999 at 12:00:01PM -0500, Jeremy Porter wrote:
It shouldn't if you router is configured to not toggle the interface down or you are using the loop-back as update source thing. I don't know for sure if 50ms is enough to cause the interface to detect it being down or not. I suspect it depends a lot on the interface.
In message <E11bRM2-000II7-00@rip.psg.com>, Randy Bush writes:
50ms is fast? You'd drop sevral packets I'd guess.
easily survivable. but will bgp flap?
--- jerry@fc.net Freeside/ Insync Internet, Inc.| 512-458-9810 | http://www.fc.net #include <sys/machine/wit/fortune.h>
-- Jared Mauch | pgp key available via finger from jared@puck.nether.net clue++; | http://puck.nether.net/~jared/ My statements are only mine. END OF LINE |
The problem becomes that if your APS gear is not properly tested and configured you can have an outage when this happens. Most people can't trust the telco to perform this configuration of their gear, so yes, in cases it would be better to get a diverse circuit path, or bury your own fiber and use your own APS gear. One thing that would work well is if the telcos would regularly test their equipment and send out an outage message saying "We will be testing our sonet/aps gear on your ckts a,b,c,d on mmddyyyy from 3am-6am to insure that we have done our job right and that the gear is still functional." I would not be opposed to the telcos sending me outage messages such as that in the least, and would love it if someone did that. I suspect though that they are spending so much time keeping up with demand that doing proactive work such as this would be almost unspeakable, but I do think that raising the bar like this would be of great benifit to them, and would help reduce the amount of scrambing when there is really a fiber cut. - jared On Wed, Oct 13, 1999 at 09:34:18AM -0700, Randy Bush wrote:
50ms is fast? You'd drop sevral packets I'd guess.
easily survivable. but will bgp flap?
-- Jared Mauch | pgp key available via finger from jared@puck.nether.net clue++; | http://puck.nether.net/~jared/ My statements are only mine. END OF LINE |
On Wed, 13 Oct 1999, Randy Bush wrote:
50ms is fast? You'd drop sevral packets I'd guess.
easily survivable. but will bgp flap?
Not unless the interface flaps and you have fast-neighbour-fallover turned on.
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Alex Rubenstein, alex@nac.net, Chief Technology Officer Net Access Corporation, 9 Mt. Pleasant Tpk., Denville, NJ 07834 Don't choose a spineless ISP; we have more backbone! http://www.nac.net -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Could someone from Verio please contact me privately? This is in regards to stop announcing a certain /24. Thank you, Michael...
Sam Thomas wrote:
semi-operational: in the event that path inside a SONET ring switches from active to protect and back, etc...does this generally show up on a a router as a series of errors, an interface bounce, or not at all?
Often both. The switch is spec'd at 50 milliseconds. Hardly a twitch in human voice perception, but a heck of a lot of bad PPP checksums in machine perception. Worse, in my miserable experience, the APS would flip back and forth. It's not supposed to, but it has. I firmly recommend: do not pay for APS. Put the money toward diverse paths. As usual, Paul Vixie has got it right! WSimpson@UMich.edu Key fingerprint = 17 40 5E 67 15 6F 31 26 DD 0D B9 9B 6A 15 2C 32
At 12:07 PM 10/13/1999 -0400, William Allen Simpson wrote:
Worse, in my miserable experience, the APS would flip back and forth. It's not supposed to, but it has. I firmly recommend: do not pay for APS. Put the money toward diverse paths. As usual, Paul Vixie has got it right!
APS is not intended to be a standalone protection factor. Implemented on routers, it will only protect you up to the first LTE. In most cases, this will be an in-house cross connect. Don't confuse APS with 1+1 protection. -Steve
At 14:27 13/10/99 +0000, you wrote:
semi-operational: in the event that path inside a SONET ring switches from active to protect and back, etc...does this generally show up on a a router as a series of errors, an interface bounce, or not at all?
in the case of you using a circuit (T1, DS3, ATM, ..) which your carrier transports over SONET, the best you'll see is a lost packet or ten. (the actual switchover time is fast; the amount of data lost is dependent on the distance of the hop -- light 'in motion' is not recoverable). in the case of you using a POS interface or similar, you may get APS notification, providing the ring that has failed is on the same element as you. cheers, lincoln.
participants (15)
-
Alex Bligh -
Alex Rubensteinc -
Charles Sprickman -
Ehud Gavron -
Eric Germann -
Jared Mauch -
Jeremy Porter -
Lincoln Dale -
michael -
Michael Heller -
Mike Heller
-
Randy Bush -
Sam Thomas -
Steve Meuse -
William Allen Simpson