The issue is on Vmware's side: VSwitch port group settings under "teaming and failover" "route based on IP Hash" must be selected.
↧
Re: Performance issues withJuniper EX4300MP
↧
Re: EX 4300 many to many mirroring
Thanks for the link.
Just wished to understand, if we can use same input port for both of the instances here?
Or can i have 1 instance with many inputs and two recorder output (using no-mac learning) i guess.
↧
↧
EX4500 dropping packets, will EX4600 do the same thing?
When we got our new servers up and running last September we ended up with a Veeam problem during replication. I ended up opening tickets with Veeam and VMWare since replication data gets pushed through the ESXi management interface. After two months of pulling logs, analyzing running jobs, and other troubleshooting steps we found the culprit, it is the EX4500. With our old Bladecenter and old Veeam proxy server only some much data was pumped through the EX4500. With the new servers and especially the new Veeam proxy server (2 CPUs with 16 cores each for a total of 32 cores), the amount of data we were pumping through the EX4500 increased significantly during replication. The EX4500 was dropping a lot of packets, so many that the connection through the ESXi management interface during replication was interrupted. The Veeam replication job would simply bomb. The work around is simple. Within the Veeam software you can tell it the max concurrent tasks. Best practice is to make this the number of cores available. We had "Max concurrent tasks" set to 32 since the new Veeam proxy server has 32 cores. We ended up lowering it to 10 so we were not overwhelming the EX4500. The big question is will the EX4600 exhibit the same behavior? I am not sure if this is dictated by the buffer on the switch, the CPU the switch uses, or other characteristics of the switch.
Any idea if the EX4600 will be better able to handle this flood of data across a 10gig connection when replication jobs run?
↧
Re: EX4500 dropping packets, will EX4600 do the same thing?
This is generally dictated by the buffers. I see you had a previous thread here about the EX4500. There is a method to change how the shared buffers work on EX products
set class-of-service shared-buffers
By default, EX4500 allocated buffers on a per interface, whether interface was active or not. This could then create overflow in default 'best-effort' queue (assuming no other COS/QOS set).
EX4600 does function differently and does have greater buffer size (EX4500 smalled, EX4550 bigger, EX4600 biggest). I know EX4600 is 12M shared, not sure of numbers for others. In general have not only seen issues with EX4500, not with EX4550 or EX4600. For more info on how to configure (if default is no good) shared buffers on EX4600 see here - https://www.juniper.net/documentation/en_US/junos/topics/example/cos-shared-buffer-allocation-lossless-qfx-series-configuring.html
Note the page references QFX, but EX4600 and QFX5100 have the same hardware and buffer implentation, so in this case equally applies to EX4600. You should be fine with EX4600.
HTH
↧
push tag with l3 interface on client's QnQ
--------------- --------------- ---------
+ + xe-0/0/41 + + 100G + +
+ 3750x-48 +-------------+ qfx5110-48s +---------+ mx204 +
+ + + + + +
--------------- --------------- ---------
| |xe-0/0/1
| |
| |
client qnq client qnqCan QFX or MX push tag with l3 interface (for BGP) into client's QnQ? Now we could configured BGP on mx204 in native vlan only.
Client have QnQ between 10G ports on 3750x and qfx5110.
QFX:
### client port ###
interfaces {
xe-0/0/1 {
flexible-vlan-tagging;
mtu 9216;
encapsulation extended-vlan-bridge;
unit 0 {
vlan-id-list 1-4094;
input-vlan-map push;
output-vlan-map pop;
}
### vlans conf ###
vlans {
vl311 {
interface xe-0/0/41.311;
interface xe-0/0/1.0;
interface et-0/0/48.311;
}
### port to mx204 ###
interfaces {
et-0/0/48 {
flexible-vlan-tagging;
mtu 9216;
encapsulation flexible-ethernet-services;
unit 311 {
encapsulation vlan-bridge;
vlan-id 311;
}
MX204:
interfaces {
et-0/0/0 {
flexible-vlan-tagging;
mtu 9216;
encapsulation flexible-ethernet-services;
gigether-options {
no-flow-control;
}
unit 311 {
vlan-id 311;
family inet {
address 10.10.10.1/31;
}
↧
↧
Persistent Mac Learning - Junos requirements
Hello all you wonderful people!
I'm having to try to configure persistent mac learning on my EX3300 switches and well I'm not sure the Junos I am currently running supports it (or I just can't find the right command) We are currently running [12.3R9.4] - and when I try to enter the set interface {interface number} mac-limit command I don't have the mac-limit as an option 
It appears that it should be available - but I can't configure it.
So is this something that has to be configured on the SRX router not the EX switch?
And yes - I am *very* confused.
Thanks in advance!
Lirria
↧
Re: Persistent Mac Learning - Junos requirements
Hello,
Please try under give hierarchy
set ethernet-switching-options secure-access-port interface ge-0/0/0.0 mac-limit 1
set ethernet-switching-options secure-access-port interface ge-0/0/0.0 mac-limit action log
set ethernet-switching-options secure-access-port interface ge-0/0/0.0 persistent-learning
set ethernet-switching-options storm-control interface ge-0/0/0.0
Mark my response as solution if its solves your problem, kudos are appreciated too!
Thanks
Vishal
↧
Re: Persistent Mac Learning - Junos requirements
Yup that's it - thank you - I do feel rather stupid because about 5 minutes ago I stumbled into the
set ethernet-switching-options - (you don't want to know just how long I searched for it - let's just call it a blonde day 
)
Really appreaciate the response to this.
Have a wonderful day!
Lirra
↧
Re: EX4500 dropping packets, will EX4600 do the same thing?
Thanks for the assistance. Besides the thread that I started quite some time ago about the EX4500, I found a few other references where people were having issues with dropped packets on this switch. It sound like the EX4600 with bigger buffers and if need be allocating different per the article you linked to should take care of this issue. I would certainly like to be able to crank up "Max concurrent tasks" in Veeam to make best use of my Veeam proxy server and signicantly cut my backup and replication times.
↧
↧
Re: push tag with l3 interface on client's QnQ
Hello,
wrote:
Can QFX or MX push tag with l3 interface (for BGP) into client's QnQ? Now we could configured BGP on mx204 in native vlan only.
It can but You have to know the customer C-VLAN tag value - MX or QFX cannot "hunt" for a C-VLAN to figure out which exact C-VLAN carries BGP traffic.
Assuming C-VLAN tag for BGP is 411, please see below example MX204 config:
set interfaces et-0/0/0 unit 411 vlan-tags outer 311 inner 411 set interfaces et-0/0/0 unit 411 family inet address 203.0.113.1/24
What will happen on et-0/0/0.411 subinterface is:
- when a dual-tagged Ethernet frame with innner VLAN tag 411 and outer VLAN tag 311 enters et-0/0/0, both tags will be popped and underlying IP or ARP packet examined
- when there is an IP or ARP packet that needs to go out of et-0/0/0.411, two VLAN tags will be pushed (inner 411 outer 311), and resulting dual-tagged Ethernet frame forwarded out of et-0/0/0.
HTH
Thx
Alex
↧
QFX5100 port lockup
Hi!
We have a few QFX5100s and in one of them four ports do not accept QSFP+ DAC cables (and probably not fiber QSFP+ modules) right now. The ports in question are et-0/0/16 to 19. We could reboot the switch and the ports would probably come up, but as this is a production switch, we'd like to explore other options first. I suspect that the QFX5100-24Q has its physical ports divided into port groups where a certain number of ports share some hardware, like a PHY or MAC chip. Reading up on the matter, I see that the QFX10k 36Q has its ports in groups of three, as do lots of other 40/100 G switches. In a deep dive PDF about the QFX5100, I read that the 24Q version can channelize ports 4 to 24, not 0-3. That makes me think that this model has ports grouped in groups of four. That would align with ports 16-19 that we're having problems with. Does anyone know if this is indeed so, and can this common piece of hardware (4-port MAC/PHY/whatever) be reset without too much risk of upsetting the rest of the system?
I have seen a similar scenario in an Avaya VSP9k where ports were divided in groups of 8 and each group consisted of two lanes with 4 ports each. One lane (ports 21-24) locked up and caused a massive traffic storm as the combined MC-LAG ICL/ICCP link (called IST in Avayas SMLT terminology) was on port 23 and 24.
From other vendors (Avaya, Extreme, Cisco and others) I know port grpoups exist, but I seem to be unable to find info on the QFX5100-24Q in that regard. Can anyone help?
/Fredrik
↧
Re: QFX5100 port lockup
...and by the way, we can see the DAC cables when doing this:
> start shell> vty fpc0> show qsfp list
but not when typing "show chassis hardware".
When inserting a DAC in port 16, we get the following:
Oct 19 12:33:42 qfx5100-1 fpc0 qsfp-0/0/16 plugged in
Oct 19 12:33:42 qfx5100-1 fpc0 FAILED to write the QSFP IDEEPROM for port: 16 start addr: 127 - length: 1
Oct 19 12:33:42 qfx5100-1 fpc0 qsfp_tk_select_mem_page: qsfp-0/0/16 failed to select mem page 0 addr 0x7f err 1000
But, again, I'm mostly interested in if the fact that all ports work except ports 0/0/16 to 19 can be due to some port grouping, and of course if we can recover without rebooting the switch or pfe.
↧
Re: EX3300 [12.3R3.4] returns an error: remote unlock-configuration failed
Good Day Dawid,
I have a pretty obvious suggestion - to upgrade Virtual-Chassis to one of the latest releases from official Juniper download web-page.
JunOS which you are running is relatively old - release date is 18-Sep-2013.
Similar issue was observed on such old releases and has been fixed in last ones.
We do not have enough data to clarify is it a exact match, but still... my advice will be to upgrade JunOS as it already resolved similar cases before.
Thank you!
↧
↧
Re: EX3300 [12.3R3.4] returns an error: remote unlock-configuration failed
Hello again,
As I can't perform the switch restart, I plan to copy necessary files from working one.
The question is: exactly which files in which directory shall I copy from working switch to damaged one?
Can I just use:
file copy fpc0:/path/file fpc1:/path/file
?
Files on affected switch:
drwxrwxr-x 2 root wheel 512 Dec 24 2011 .snap drwxr-xr-x 4 root wheel 512 Jul 1 13:19 db -rw-r----- 1 root wheel 3666 Jul 1 12:40 juniper.conf.1.gz -rw-r----- 1 root wheel 3666 Jul 1 12:06 juniper.conf.2.gz -rw-r----- 1 root wheel 3665 Jul 1 12:00 juniper.conf.3.gz -rw-r----- 1 root wheel 3666 Jul 1 12:58 juniper.conf.gz -r-s------ 1 root wheel 32 Jul 1 13:14 juniper.conf.md5 -rw------- 1 root wheel 672 Jul 1 12:59 ssh_host_dsa_key -rw-r--r-- 1 root wheel 601 Jul 1 12:59 ssh_host_dsa_key.pub -rw------- 1 root wheel 227 Jul 1 12:59 ssh_host_ecdsa_key -rw-r--r-- 1 root wheel 173 Jul 1 12:59 ssh_host_ecdsa_key.pub -rw------- 1 root wheel 976 Jul 1 12:59 ssh_host_key -rw-r--r-- 1 root wheel 641 Jul 1 12:59 ssh_host_key.pub -rw------- 1 root wheel 1675 Jul 1 12:59 ssh_host_rsa_key -rw-r--r-- 1 root wheel 393 Jul 1 12:59 ssh_host_rsa_key.pub -rw-r--r-- 1 root wheel 1836 Jun 14 2013 usage.db drwxr-xr-x 2 root wheel 512 Jul 1 13:19 vchassis root@st_0-1:LC:1% cd db root@st_0-1:LC:1% ls -l total 48 -rw-r--r-- 1 root wheel 1684 Jul 1 12:58 commits drwxr-xr-x 2 root wheel 1536 Jul 1 12:58 config -rw-r--r-- 1 root wheel 15879 Jul 1 13:18 dcd.snmp_ix drwxr-xr-x 2 root wheel 512 Apr 11 2000 scripts -rw-r--r-- 1 root wheel 99 Jul 1 12:58 snmp_engine.db root@st_0-1:LC:1% cd config/ root@st_0-1:LC:1% ls -l
Dawid
↧
Re: EX3300 [12.3R3.4] returns an error: remote unlock-configuration failed
The active configuration is /config/juniper.conf.gz
copy the known good one to the failing member
↧
Recommended replacement switches
We have a VC consisting of 2 x EX4200-48t and 2 x EX4500-40f.
What would be considered replacement hardware for this chassis.?
thanks
↧
Re: Recommended replacement switches
2x EX4300-48T + 2x EX4600-40F. 24 ports are SFP+, you will need 4x 40Gb->10Gb breakout cables or two 8-port expansion modules for the other 16.
or
2x EX4300-48T + 2x QFX5100-48S
↧
↧
Re: Recommended replacement switches
If you are goiong the QFX5000 Series route vs EX4600, even though EX4600 and QFX5100 are the exact same hardware, difference is SW, I would suggest going with QFX5110 vs QFX5100 - has all plus more. If any product is nearing a point of becoming End of Sale, it would be QFX5100 as QFX5110 has been a better option for a long time, EXCEPT for QFX5100-48T. In 2020 when QFX5120-48T becomes an 10GBase-T option, then End of Sale of QFX5100 may occur sometime after.
↧
Re: Recommended replacement switches
OK, both suggestions are good, thanks both of you for replying
↧
Re: Recommended replacement switches
AFAIK you cannot VC EX4300's with QFX5110's.
↧