Archives
Archives
| From: | karl@××××××××.se | ||
|---|---|---|---|
| To: | gentoo-user@l.g.o | ||
| Subject: | Re: [gentoo-user] Network switch - LED will not turn ON | ||
| Date: | Tue, 27 Apr 2021 09:04:06 | ||
| Message-Id: | 20210427090358.23FDA83D15CE@turkos.aspodata.se | ||
| In Reply to: | Re: [gentoo-user] Network switch - LED will not turn ON by Wols Lists |
||
| 1 | Wol: |
| 2 | > On 26/04/21 22:53, thelma@×××××××××××.com wrote: |
| 3 | > > If they cable is the problem the LED port on the 70ft long cable |
| 4 | > > on the switch would be orange "not green"; correct me anybody if |
| 5 | > > I'm wrong. |
| 6 | > |
| 7 | > I guess it depends what's wrong with the cable. And what the green light |
| 8 | > is testing for. |
| 9 | |
| 10 | As can be seen from |
| 11 | |
| 12 | https://en.wikipedia.org/wiki/Autonegotiation#Electrical_signals |
| 13 | |
| 14 | autonegotiation only runs at 10Base-T speeds. Its use is to make |
| 15 | sure that the endpoints are operating with the same parameters. |
| 16 | |
| 17 | Note, it does not tests whether the cable and connector are suitable |
| 18 | for that negotiated link speed. |
| 19 | |
| 20 | Also, |
| 21 | |
| 22 | https://en.wikipedia.org/wiki/Gigabit_Ethernet#1000BASE-T |
| 23 | |
| 24 | says that the tests are not done on all cable pairs, so even |
| 25 | if pair 1 and 2 are perfect, faulty pair 3 and 4 may make |
| 26 | communication fail. |
| 27 | |
| 28 | /// |
| 29 | |
| 30 | So if the link led lights up, but the communication fails, |
| 31 | something between the mii's (the electronics, the chips) |
| 32 | |
| 33 | https://en.wikipedia.org/wiki/Media-independent_interface |
| 34 | |
| 35 | is faulty and it is with high probability that the cable is faulty. |
| 36 | |
| 37 | Regards, |
| 38 | /Karl Hammar |