What you find? Jaime Solorza On Mar 31, 2015 3:33 PM, <[email protected]> wrote:
> Seen this happen before. Has to be a potential between one of the > ethernet lines and ground. Sometimes it is because someone used a LV > version instead of a HV version. > > *From:* Jaime Solorza <[email protected]> > *Sent:* Tuesday, March 31, 2015 11:13 AM > *To:* Animal Farm <[email protected]> > *Subject:* Re: [AFMUG] Ethernet Interference from Grounding using GIGE-APC > > > Bring out the Holy Hand ...oops bring out the voltmeter and see what you > get. Compare to non affected lines. Ground isolation is a science all to > itself > > Jaime Solorza > On Mar 31, 2015 11:04 AM, "Sean Heskett" <[email protected]> wrote: > >> if there is voltage on the ground then yes it will cause ethernet >> errors. >> >> >> >> On Tue, Mar 31, 2015 at 10:49 AM, Nate Burke <[email protected]> wrote: >> >>> I just changed out an ATOM mikrotik to a CCR, and found that it is >>> Incrementing FCS Errors exactly every 30 seconds on the Ethernet line from >>> a Trango LYNX. The lynx is powered via DC, so the Data line is clean of >>> POE. I am using a GIGE-APC Unit on the Ethernet line. >>> >>> What I found through testing, was that the FCS Errors only happened when >>> the GIGE-APC was plugged into the chassis. They happened at exact 30 >>> second intervals. Because of this I'm guessing that there is something >>> introduced via the ground that is causing them. If I pull the GIGE-APC out >>> of the Chassis, the errors stopped, if I plugged it back in, the errors >>> came back, at the exact same clock points (in my CCR Log, always at 06 and >>> 36 seconds regardless of the timing of plugging/unplugging). I'm using the >>> 1U APC chassis, and there are other GIGE-APC, 444APC-HV, and GIGE-POE-APC >>> units in it. None of the other equipment is plugged into the CCR, so unsure >>> if it would increment FCS Errors on those. They are all plugged into >>> RB1100's, and have no errors logged in the mikrotik. >>> >>> Any place to start troubleshooting something like this? Would it make >>> sense that something on the ground plane could cause these Ethernet errors? >>> >>> I believe that this is causing some very low level packetloss through >>> the link, about 1 packet every 5 minutes or so, I'm guessing based on if >>> the ping falls directly on the error frame. >>> >>> Nate >>> >>> >> >
