Just a quick question to those using juniper MX series routers.. or perhaps any series - is there a known issue with the console port dying ? We've seen this happen on two different boxes - completely unresponsive after running for a few weeks, reboot doesn't fix it, and no output at all on it, even during the freebsd boot sequence. So before I start second and third-guessing what I've been doing to it, is this a known issue that people have seen elsewhere ?
Never seen it myself.. what does JTAC say? Does your cable work on other similar boxes? Just to rule out a bad cable/pinout etc.. On 6/06/2012, at 2:12 PM, Ian Batterbee wrote:
Just a quick question to those using juniper MX series routers.. or perhaps any series - is there a known issue with the console port dying ? We've seen this happen on two different boxes - completely unresponsive after running for a few weeks, reboot doesn't fix it, and no output at all on it, even during the freebsd boot sequence.
So before I start second and third-guessing what I've been doing to it, is this a known issue that people have seen elsewhere ? _______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog
We have 12 MX80's and they've been running fine for a year so far. Had a
couple with disk corruption after reboots/upgrades (which were easy enough
to fix with rescue USB drives), although tbh we don't console that much -
we have out-of-band set up to them but always login via ssh
On Wed, Jun 6, 2012 at 2:12 PM, Ian Batterbee
Just a quick question to those using juniper MX series routers.. or perhaps any series - is there a known issue with the console port dying ? We've seen this happen on two different boxes - completely unresponsive after running for a few weeks, reboot doesn't fix it, and no output at all on it, even during the freebsd boot sequence.
So before I start second and third-guessing what I've been doing to it, is this a known issue that people have seen elsewhere ?
_______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog
-- Sam Russell Network Operations Research & Education Advanced Network NZ Ltd ddi: +64 4 913 6365 mob: +64 21 750 819 fax: +64 4 916 0064 http://www.karen.net.nz
Nope - we have quite alot of MX boxen in the network and havent seen a
console port die yet...
Have you asked JTAC? Is it a common console router/connector/laptop/usb
serial thing?
On Wed, Jun 6, 2012 at 2:12 PM, Ian Batterbee
Just a quick question to those using juniper MX series routers.. or perhaps any series - is there a known issue with the console port dying ? We've seen this happen on two different boxes - completely unresponsive after running for a few weeks, reboot doesn't fix it, and no output at all on it, even during the freebsd boot sequence.
So before I start second and third-guessing what I've been doing to it, is this a known issue that people have seen elsewhere ?
_______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog
-- Callum Barr me(a)callumb.com
Havent come across this with console server to MX series connections. I'd be suspicious of a usb-to-serial adapter if that's in the equation anywhere. Brendan Baker Opengear On 06/06/2012, at 12:31 PM, Callum Barr wrote:
Nope - we have quite alot of MX boxen in the network and havent seen a console port die yet...
Have you asked JTAC? Is it a common console router/connector/laptop/usb serial thing?
On Wed, Jun 6, 2012 at 2:12 PM, Ian Batterbee
wrote: Just a quick question to those using juniper MX series routers.. or perhaps any series - is there a known issue with the console port dying ? We've seen this happen on two different boxes - completely unresponsive after running for a few weeks, reboot doesn't fix it, and no output at all on it, even during the freebsd boot sequence. So before I start second and third-guessing what I've been doing to it, is this a known issue that people have seen elsewhere ?
_______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog
-- Callum Barr me(a)callumb.com _______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog
Have a few Juniper boxes of various flavours under my purview and have never had this issue. Have had a couple of lock-outs but a reboot fixed it.
Cheers
Wayne
On 06/06/2012, at 12:31 PM, Callum Barr wrote:
Nope - we have quite alot of MX boxen in the network and havent seen a console port die yet...
Have you asked JTAC? Is it a common console router/connector/laptop/usb serial thing?
On Wed, Jun 6, 2012 at 2:12 PM, Ian Batterbee
Thanks for the replies.
To answer questions, this is a real db9 port on a linux laptop, no usb
trickery involved, and besides, the same cable works fine when moved to
another mx80, or indeed to the aux port of the mx80 with the problem.
We RMA'd another one with identical symptoms at JTAC's request - this is
the replacement, which worked fine for about 2 weeks, then was untouched
for a few more weeks, and then today I found the console port dead - exact
same symptoms as before, and it seems rather too odd to be a coincidence.
Hence my question about whether this is a common thing.
I'll follow up with JTAC.
On 6 June 2012 14:12, Ian Batterbee
Just a quick question to those using juniper MX series routers.. or perhaps any series - is there a known issue with the console port dying ? We've seen this happen on two different boxes - completely unresponsive after running for a few weeks, reboot doesn't fix it, and no output at all on it, even during the freebsd boot sequence.
So before I start second and third-guessing what I've been doing to it, is this a known issue that people have seen elsewhere ?
Possibly your laptop or cable is killing it. Rs323 cabling and ports use quite high voltage for signalling in comparison to say USB or ethernet... If the chip set of the Linux box is borked you may be frying the port. Despite what others have said I would try a USB serial adaptor (I suggest fddi chip based ones as found at dick smiths) to use instead of a 'real' port. These tend to have much better voltage regulation in them than ancient laptops.
On Wed, 6 Jun 2012, Joel Wiramu Pauling wrote:
Possibly your laptop or cable is killing it. Rs323 cabling and ports use quite high voltage for signalling in comparison to say USB or ethernet... If the chip set of the Linux box is borked you may be frying the port.
Possible, but only if console ports are themselves sub-spec. Generating enough voltage to fry a to-spec RS232 port is beyond the wildest dreams of most PCs; they simply don't have the electronics to generate the voltages needed to cause damage. RS232 requires that inputs tolerate at least ±25 V, but sense high/low transitions somewhere in ±3 V. (And outputs should likewise be immune to indefinite short-circuit to any voltage in ±25 V.) A "nice" RS232 receiver will operate with threshold at around +2 V (thus being being compatible with RS422 driving at 0/+5 V), but won't blow even with any pin connected to ±47 V (thus being safe for telco work / abuse). Good luck finding a new device that'll meet that improved spec... -Martin
participants (8)
-
Brendan Baker
-
Callum Barr
-
Ian Batterbee
-
Joel Wiramu Pauling
-
Martin D Kealey
-
Nathan Ward
-
Sam Russell
-
Wayne Kampjes