Quantcast
Channel: All Ethernet Switching posts
Viewing all 10307 articles
Browse latest View live
↧

VCP status on QFX5110-48S Virtual Chassis

December 24, 2018, 8:58 pm
$
0
0

Hello community,

 

I am required to do two-members virtual chassis on QFX5110-48S with single 100GE port. Before put it into production, I am testing with single 40GE port instead.  I notice that Trunk ID is -1.

 

> show virtual-chassis vc-port
localre:
--------------------------------------------------------------------------
Interface Type Trunk Status Speed Neighbor
or ID (mbps) ID Interface
PIC / Port
0/48 Configured -1 Up 40000 1 vcp-255/0/48

fpc1:
--------------------------------------------------------------------------
Interface Type Trunk Status Speed Neighbor
or ID (mbps) ID Interface
PIC / Port
0/48 Configured -1 Up 40000 0 vcp-255/0/48

 

Questions are:
1. Is it ok and traffic working normally with -1 value?

2. Is it ok to do with single 100GE for VPC?

 

Thanks and regards,

↧
Search

Re: VCP status on QFX5110-48S Virtual Chassis

December 25, 2018, 3:37 am
$
0
0

1. Trunk ID of -1 defines that no VCP LAG trunk is up. More information about this here: https://www.juniper.net/documentation/en_US/junos/topics/reference/command-summary/show-virtual-chassis-vc-port.html

 

2. I would never deploy a virtual chassis with only one VCP link. If this link goes down the two switches both goes into a master state (if configured with no-split-detection per recommendation operating two-member virtual chassis). If you really need the 100G ports, I would suggest to implement 8-10 x SFP+ links for VCP connections to avoid being hit with a defunct enviroment due to a single connection going down.

↧
↧

Re: How to limit bandwidth on the interface of ethernet switch EX3200

December 25, 2018, 8:57 am
↧

Re: random high latency issues on ex2300 series switches

December 26, 2018, 4:00 am
$
0
0

Thanks for the response, that is one thing I'm trying to keep an eye on.

Do you know of a way to get more of a continous update on resource useage, like TOP? 

↧

Re: random high latency issues on ex2300 series switches

December 26, 2018, 3:02 pm
↧
↧

unrequested reboot of EX2300

December 27, 2018, 2:17 am
$
0
0

Hi, we have three EX2300 switches (2x 48T , 1x 24P). Today all three switches performed an unrequested reboot, all of them within only some minutes. It was the second time that this happened.

All switches are running "18.3R1.9",

All show "Last reboot reason 0x20000Smiley Frustratedwizzle reboot".

All show "% sysctl hw.re.reboot_reason

hw.re.reboot_reason: 131072".

Any ideas what is going wrong here?

Thank you for your help
Regards

↧

1GE copper on QFX5110-48S

December 27, 2018, 3:25 am
$
0
0

Hello,

 

I read some documents and still not be able to find a good understanding whether QFX5110-48S support 1GE copper?

 

Appreciate any input.

↧

Re: 1GE copper on QFX5110-48S

December 27, 2018, 3:59 am
$
0
0

They are supported. Can be figured out via Pathfinder -> Hardware compatibility tool (direct link to the requested information: https://apps.juniper.net/hct/category/#catKey=100001&modelType=All&pf=QFX+Series )

 

NOTE! It's not supported to have the 1G copper transceivers installed in the same column of ports as the RJ45 locking mechanism on these will get stuck to each other without any way to remove them from the SFP+ ports. Usually I decide that copper transceivers should go into either even or uneven ports - not a mix.

 

 

↧
Search

Re: random high latency issues on ex2300 series switches

December 27, 2018, 7:45 am
$
0
0

Yeah, I was looking for something that isn't static, like how TOP continously updates those fields.

Maybe that isn't an option with this command. 

↧
↧

Re: unrequested reboot of EX2300

December 27, 2018, 8:29 am
$
0
0

Just a wild shot in the dark.. Could heat have been the issue ?   See URL as to why im asking / guessing.

Are they all in  a VC ?

What was the uptime on them before ?

show system core-dumps

 

https://forums.juniper.net/t5/Junos/Unusual-Reboot-Reason-on-Juniper/td-p/309913

Code 0 means reboot due to power failure or power cycle.  
Code 1 is  is a reboot triggered by hardware watchdog. 

 

Also seen on an old maillist archive .. not sure it has anything extra to give but ever bit helps..


sysctl hw.re.reboot_reason

The model of routing-engine will determine how interesting this is. If you have 0, it means the reason is unknown.

If you have anything other than 0, send me the output of 'show chassis hardware' and the value.

 

Running that rev on stack of 3400s 9 members deep for a while now and nothing like that..

 

If you dont for some reason want to do push this to jtac .. would you mind just going through the motions and

collecting the data anyways.. This way when someone smarter than I get on the list the data is still there to poll from.

 

https://kb.juniper.net/InfoCenter/index?page=content&id=KB26813&actp=METADATA

↧

Difference between SFP-1GE-T and GLC-T/1000BASE-T SFP

December 27, 2018, 1:15 pm
$
0
0

Can anyone tell me the difference between Junper compatible SFP-1GE-T and Cisco GLC-T (or Generic IEEE std 1000BASE-T) SFP and why? Our EX2300-24P swith's SFP/SFP+ ports don't recognize GLC-T or 1000BASE-T.

↧

Re: Difference between SFP-1GE-T and GLC-T/1000BASE-T SFP

December 27, 2018, 2:34 pm
$
0
0

From a standards perspective, nothing--they're both 1000BASE-T. Third-party optics usually work, but aren't guaranteed to. Do they not show up in show chassis hardware, show interfaces, or something else?

↧

Re: Difference between SFP-1GE-T and GLC-T/1000BASE-T SFP

December 28, 2018, 3:03 am
$
0
0

We made the experience that for any reason, third party copper SFPs hardly work on Juniper devices, with different problem states (e.g. physical link is up, even there is no cable connected).

 

If you need reliable copper connections, I can just recommend you to use original Juniper copper SFPs. For EX2300 platform, these would be:

 

EX-SFP-1GE-T
QFX-SFP-1GE-T
SRX-SFP-1GE-T

 

Cheers

↧
↧

Re: unrequested reboot of EX2300

December 29, 2018, 8:46 am
$
0
0

Look to see if there are any core dumps from the unexpected reboot.  These can be uploaded to a JTAC case to determine the reason for the problem.

 

show system core-dumps

 

↧

Re: random high latency issues on ex2300 series switches

December 29, 2018, 8:48 am
$
0
0

You can drop to the shell and run top directly too.

 

start shell

 

↧
Search

Moved uplink module to a different switch won't work

December 30, 2018, 11:10 am
$
0
0

Hello,

Basically what i'm trying to do is replace a couple of EX3200 edge switches, each with their own fibre uplink module, with a stacked pair of EX4200s thus reducing the number of fibre connections to the core switch from 2 to 1. I have take the uplink module out of one of the EX3200s and put it into the master EX4200. I created a new IP address for accessing the new switch on vlan interface unit 0 and gave it a new host name. 

 

I just can't get any IP connectivity to the core switch, nor from the core switch to the new edge switch. They do see each other as LLDP neighbours though. 

 

As on the old switch I set a static route to the core using the routing-options command. 

I also tried moving the SFP/fibre connector to a different slot on the uplink module.

I made sure the link is set as trunk, with all vlans tagged and default as the native VLAN all on both ends of the connection exactly the same.

I made sure the speed and duplex are the same and the link negotiation has completed. 

Rebooted the switch.

Nothing seems to work. When I ping the core I get 100% packet loss and likewise from the core to the new edge switch. I can ping the IP of the new switch locally though just fine.

 

However as soon as I put the uplink module back into the old switch it works just fine.

Happy to provide any further details.

Any suggestions would be greatly appreciated!

 

↧

Re: Moved uplink module to a different switch won't work

December 30, 2018, 1:11 pm
↧
↧

QinQ Tunneling on QFX5110

December 31, 2018, 1:59 am
$
0
0

Hello,

 

I have read some documents but I could not find a good solution to make QinQ tunneling work on QFX5110-48s with Junos 17.3R3.10. The attached file is the sample network connectivity.

 

Appreciate for any advices or example configuration.

Thanks and regards,

 

↧

Re: QinQ Tunneling on QFX5110

December 31, 2018, 1:26 pm
↧

QFX-5100 DHCP Relay Broken under 18.4?

December 31, 2018, 1:39 pm
$
0
0

I've upgraded a pair of QFX5100s to run 18.4R1, and DHCP relaying is now broken...  I rolled-back to 18.3 on one of them, and the previous functionality was restored.  Looks like new functionality with "tenants" was implemented, but that appears to break the base functionality.

 

Looking to see if anyone else is seeing this, and to give a heads-up to Juniper Support.  PM me for a PCAP. 

↧
Viewing all 10307 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>