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 Chriss Miller on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
3com superstack 3300
- Thread starter ManMan
- Start date
First, check the rev level of the firmware.. 3com had known issues of high error counts( Tx buffers/Rx buffers). If I'm not mistaken, 3.x is required. Also, if the device is capable of full duplex, lock the ports to 10 1/2
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
- Thread starter
- #3
Wybnormal,
Thanks for your advice. I looked at 3com's web site, the latest firmware is 2.62 for superstack II 3300. I will try that after hour.
Let me explain our setup so you can picture it better. In this office, the setup is 4 superstack connected with Matrix cable to form one unit. There another 2 superstack hook up by Matrix cable to form another unit. This 2 units connected with crossover cable. We place most server in the unit 1. No particular VLAN is set up. Storm Broadcast control is Enable.
However, we have updated it from previous version to version 2.52 some time ago. Port error problem still exist.
Sometimes it reach up to 200 something when I look at the chart.
The only way I can get around it is to reboot all the switches. But after some time. It come again. Sometimes it come back in couple weeks or one month. It is just unpredictable. It is just pain of neck. We can't even finish the backup at night because of this port errror slow the backup speed from 150M/min to 6M/min.
In some of our remote office. Some cases are even worse?
2 superstack 3300 connected with Matrix cable. Interanal packets some time not able to pass through both swithes to reach the router and then the internet.
Sorry, I am slow, what exactly you mean by "lock the port to 10 1/2"
Do you think creating 2 VLAN will help? What about enable Span Tree and Tagging?
Thank you very much in advance
Thanks for your advice. I looked at 3com's web site, the latest firmware is 2.62 for superstack II 3300. I will try that after hour.
Let me explain our setup so you can picture it better. In this office, the setup is 4 superstack connected with Matrix cable to form one unit. There another 2 superstack hook up by Matrix cable to form another unit. This 2 units connected with crossover cable. We place most server in the unit 1. No particular VLAN is set up. Storm Broadcast control is Enable.
However, we have updated it from previous version to version 2.52 some time ago. Port error problem still exist.
Sometimes it reach up to 200 something when I look at the chart.
The only way I can get around it is to reboot all the switches. But after some time. It come again. Sometimes it come back in couple weeks or one month. It is just unpredictable. It is just pain of neck. We can't even finish the backup at night because of this port errror slow the backup speed from 150M/min to 6M/min.
In some of our remote office. Some cases are even worse?
2 superstack 3300 connected with Matrix cable. Interanal packets some time not able to pass through both swithes to reach the router and then the internet.
Sorry, I am slow, what exactly you mean by "lock the port to 10 1/2"
Do you think creating 2 VLAN will help? What about enable Span Tree and Tagging?
Thank you very much in advance
The "lockup" should be on the server side and switch side. Not all NICs when they are 10/100 work quite right so the best is to lock in the speed to 10 1/2 or whatever it can do rather then let autoneg do it's thing.
We lock down the ports on the switches to 10 1/2 for workstations, 100 full for servers. The auto-negotiation is turned OFF on all ports. We put the lateset firmware updates on the switches and this mimized the errors. Didnt get rid of them completely but it's MUCH less then it has been. We see the problems on ports with heavy loads.
Which errors are you seeing? is a buffer error? collision error?
Any changes to the network? Have you checked the aggareate output of these stack switches to make sure you are pushing the limits of the ethernet connection feeding back? You might want to split the stack to reduce the load across the switches.
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
We lock down the ports on the switches to 10 1/2 for workstations, 100 full for servers. The auto-negotiation is turned OFF on all ports. We put the lateset firmware updates on the switches and this mimized the errors. Didnt get rid of them completely but it's MUCH less then it has been. We see the problems on ports with heavy loads.
Which errors are you seeing? is a buffer error? collision error?
Any changes to the network? Have you checked the aggareate output of these stack switches to make sure you are pushing the limits of the ethernet connection feeding back? You might want to split the stack to reduce the load across the switches.
Mike S
"Diplomacy; the art of saying 'nice doggie' till you can find a rock" Wynn Catlin
I think you have to manually set the switch's port duplex setting and the nic's duplex setting so they match. Especially on older pc's and servers. I've read that the switch does not autosense properly AND, when you reboot the client pc's and servers, sometimes the nics are set to AUTO, and pick whatever they think is a good connection type at the time (which may be different than before)...
I had a mixed 10/100 environment, and I replaced as many of the old nics as I could with 10/100 3Com cards. (Had to find a few EISA 10/100 cards, which was tough, but they are out there.)
I wound up semi-abandoning the concept of "autosensing"
...i.e., I put any 10mb connections into old 10mb hubs, and any 100 mb connections into 10/100 dualspeed hubs. I then put all the hubs starring out of the switches, and let the switch handle the differences. (Due to economics, we could not replace all of the hubs with switches).
That cleaned things up and the collisions disappeared.
I had a mixed 10/100 environment, and I replaced as many of the old nics as I could with 10/100 3Com cards. (Had to find a few EISA 10/100 cards, which was tough, but they are out there.)
I wound up semi-abandoning the concept of "autosensing"
...i.e., I put any 10mb connections into old 10mb hubs, and any 100 mb connections into 10/100 dualspeed hubs. I then put all the hubs starring out of the switches, and let the switch handle the differences. (Due to economics, we could not replace all of the hubs with switches).
That cleaned things up and the collisions disappeared.
Guest_imported
New member
- Jan 1, 1970
- 0
Dear All,
Yesterday had a problem with 3300 3Com stacks...
4switches connected to other 4 switches with fiber
The stacks are configured with matrix modules
The issue is after diconnecting the fiber between two floors
the Broadcast storm stops and is seen only on one wing
after reconnecting it occurs on all switches
The floor which which shows Packet transfer leds lit continuously...was seen with the switch creating the problem
after the culprit switch is disabled from the stack the storm stops.
I am sure the broadcast is coming from one of the machines connected in that switch..
The issue is if i disconnect all the machines on that switch
and re-cascade 4 switches...there is broadcast at regular intervals say about 30 seconds..
Any suggestions are most welcome...
g_1@onebox.com
"Working Together Means Winning Together"
Yesterday had a problem with 3300 3Com stacks...
4switches connected to other 4 switches with fiber
The stacks are configured with matrix modules
The issue is after diconnecting the fiber between two floors
the Broadcast storm stops and is seen only on one wing
after reconnecting it occurs on all switches
The floor which which shows Packet transfer leds lit continuously...was seen with the switch creating the problem
after the culprit switch is disabled from the stack the storm stops.
I am sure the broadcast is coming from one of the machines connected in that switch..
The issue is if i disconnect all the machines on that switch
and re-cascade 4 switches...there is broadcast at regular intervals say about 30 seconds..
Any suggestions are most welcome...
g_1@onebox.com
"Working Together Means Winning Together"
Guest_imported
New member
- Jan 1, 1970
- 0
I have a 3com 3300 Superstack II. How do I check out the errors on the ports?
Guest_imported
New member
- Jan 1, 1970
- 0
Hi i have got a 3Com Superstack 3300, it will not power up.
Do you think it could be the power unit? or if not what else could be wrong with it?
thanks
Michael
Do you think it could be the power unit? or if not what else could be wrong with it?
thanks
Michael
Guest_imported
New member
- Jan 1, 1970
- 0
I am getting given a 3com Superstack 2 Switch 3000 which doesnt power up. My first thing to look for when I get it is to open it up and see if the fuse has blown, if not then look at the PSU and test it for any output. If anyone knows any common reasons for them to not work please let me know !
Guest_imported
New member
- Jan 1, 1970
- 0
I have a 3300 and it has solid orange lights on, what does this mean and how do I get it back to normal?
Hello:
I have a small network setup with a Novell 5 server and a Windows 2000 client (among others) running through a Cisco 3300 SSII switch set at 100 meg full. File transfers through DOS are quite speedy, while transfert through the Novell 4.8 client are horribly slow, and such transfers frequently crash the server. Any Ideas on what migh be causing the problem. (I already tried setting the client side NIC at 100 full - no effect.
Thanks
I have a small network setup with a Novell 5 server and a Windows 2000 client (among others) running through a Cisco 3300 SSII switch set at 100 meg full. File transfers through DOS are quite speedy, while transfert through the Novell 4.8 client are horribly slow, and such transfers frequently crash the server. Any Ideas on what migh be causing the problem. (I already tried setting the client side NIC at 100 full - no effect.
Thanks
Hi,
I experienced a problem with 3 Com switch.
My whole network use to die down.
Previously what i use to do is restart the 3com switches.
The best thing you do is set all your switches to default factory settings and then try out and keep the settings of all the switches in the stack same.
Raj
I experienced a problem with 3 Com switch.
My whole network use to die down.
Previously what i use to do is restart the 3com switches.
The best thing you do is set all your switches to default factory settings and then try out and keep the settings of all the switches in the stack same.
Raj
- Status
Similar threads
- Locked
- Question