Does anyone know if it is posible to hard code a port on a 5406ZL switch to 1000FDx? I am trying to connect to a switch that I cannot control (Cisco 3550 with copper GBic) and the GBic is hard coded in teh hardware to 1000FDx. HP is telling me to set it to Auto 1000 and it should work. Well it is not ad as I have no influence on the other end I am stuck trying to get the Procurve to work. any ideas?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations derfloh on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
5406ZL won't set to 1000FDx
- Thread starter FVSD
- Start date
Since you are trying to connect via copper, is the cable you are using been verified to handle gigabit speeds?
CAT5e or CAT6 patch cable?
Hand-made or purchased pre-terminated?
Plugging into the Procurve GBIC port or a 10/100/1000 port?
CAT5e or CAT6 patch cable?
Hand-made or purchased pre-terminated?
Plugging into the Procurve GBIC port or a 10/100/1000 port?
- Thread starter
- #3
All ports default to “Auto”
“Auto” ports follow standard procedures to negotiate speed
and mode with port at the other end of the link:
• 10/100 UTP ports negotiate between 10 Mbps half-duplex (HDx),10 Mbps full-duplex (FDx), 100 HDx, and 100 FDx
• 10/100/1000 UTP ports negotiate between 10HDx, 10FDx,
100HDx, 100FDx, and 1000FDx
• HP Auto-MDIX negotiates TX/RX pair reversal for UTP ports,
eliminating need for crossover cables for switch-to-switch
connections
• Gigabit fiber ports operate at 1000FDx—do not negotiate speed and mode, but do auto-negotiate flow control
Again, has the cable been certified to run 1Gb/s (with a level IV tester)?
“Auto” ports follow standard procedures to negotiate speed
and mode with port at the other end of the link:
• 10/100 UTP ports negotiate between 10 Mbps half-duplex (HDx),10 Mbps full-duplex (FDx), 100 HDx, and 100 FDx
• 10/100/1000 UTP ports negotiate between 10HDx, 10FDx,
100HDx, 100FDx, and 1000FDx
• HP Auto-MDIX negotiates TX/RX pair reversal for UTP ports,
eliminating need for crossover cables for switch-to-switch
connections
• Gigabit fiber ports operate at 1000FDx—do not negotiate speed and mode, but do auto-negotiate flow control
Again, has the cable been certified to run 1Gb/s (with a level IV tester)?
- Thread starter
- #5
The cable running from the Cisco to the patch panel I did not test as it was provided by them and is a cross over. I did however try a straight through CCat6 cable that was tested with our Fluke and that was fine. I was not sure if I could use a straight through for that.
It is just irritating that Cisco has the GBic hard coded.
It is just irritating that Cisco has the GBic hard coded.
The Cisco GBIC port is a 1Gb only port, not a 10/100/1000, so there is no need for it to negotiate sine there is no half duplex Gb. The port should auto-MDIX if it needs to cross or not in regards to the crossover cable, so use the known tested straight thru cable. Are these switches within the same proximaty to each other? Also do you have access to where you can use a GBIC copper SFP instead of the autosensing 10/100/1000 ports on the HP?
- Thread starter
- #7
They are in the same building but on different floors right above each other. I have ordered an 1800 8 port one that HP tells me will hard code to 1000FDx in case I cannot get this to work. The hard part of all this is we are the hub and everytime I test this everyone else loses the connection. Next window is in a couple of days so we will try again.
You say that the switches are on different floors, and you also mentioned patch panel. You also mentioned that you had a Fluke and tested a CAT6 cable that tested fine.
Is the cable running between the floor this same cable or is this a patch cable your are just connecting into the patch panel? What I'm trying to get you to confirm, is that end-to-end (cable, jacks, patch panels, patch cables) are confirmed minimum CAT5e or greater and have been tested with a scanner that certifies it to run at 1Gb. Kinks, severe bends, excessive pulling when installed can cause all kind of issues when pushing the cable to those higher frequencies that a Level IV scanner will reveal.
I have setup many Catalyst switches and they are among the least forgiving when it comes to cabling infrastructure. This is one of the main reasons for connectivity issues at higher speeds because with the degradation of signal on a "problem" cable segment, there will be a problem with the auto-negotiation between the switches as they negotiate what speed they can talk to each other, thus because of the degraded cable, will talk at a lower common denomenator.
That's my opinion at least, and we all have them ;>)
Is the cable running between the floor this same cable or is this a patch cable your are just connecting into the patch panel? What I'm trying to get you to confirm, is that end-to-end (cable, jacks, patch panels, patch cables) are confirmed minimum CAT5e or greater and have been tested with a scanner that certifies it to run at 1Gb. Kinks, severe bends, excessive pulling when installed can cause all kind of issues when pushing the cable to those higher frequencies that a Level IV scanner will reveal.
I have setup many Catalyst switches and they are among the least forgiving when it comes to cabling infrastructure. This is one of the main reasons for connectivity issues at higher speeds because with the degradation of signal on a "problem" cable segment, there will be a problem with the auto-negotiation between the switches as they negotiate what speed they can talk to each other, thus because of the degraded cable, will talk at a lower common denomenator.
That's my opinion at least, and we all have them ;>)
- Status
Similar threads
- Locked
- Question