|
adorai posted:is your route in the same vrf as all of your other vlans? It sounds like you are testing from one vrf but your real traffic is in another. VLAN's are on the correct VRF. Example config below. vrf context DFGW ip route 0.0.0.0/0 10.10.2.2 vrf context management ip route 0.0.0.0/0 10.10.99.1 vlan 2 name Firewall vlan 100 name 1st_Floor vlan 200 name 2nd_Floor int vlan 2 description Firewall ip address 10.10.2.1/24 vrf member DFGW int vlan 100 description 1st_Floor ip address 10.10.10.1/24 vrf member DFGW no shutdown int vlan 200 description 2nd_Floor ip address 10.10.20.1/24 vrf member DFGW no shutdown int ethernet 101/1/1 switchport accèss VLAN 100 int ethernet 101/1/2 switchport accèss VLAN 200 int ethernet 101/1/48 switchport accèss VLAN 2 EDIT- Routing issue on the firewall. Working with the fw admin. the spyder fucked around with this message at 20:20 on Sep 1, 2015 |
#
?
Sep 1, 2015 18:11
|
|
|
|
| # ? Apr 19, 2024 21:30 |
|
|
Partycat posted:I have a ton of those out there and don't recall that. Should be amber on down, but green when up or maybe looped? Didn't think that was a thing on those. Maybe the LED isn't displaying status, but something else instead ? Looking at 4500X right now, with copper SFP it does nothing, with nothing it does nothing. With disabled it is amber, and loopback, it is nothing. No idea why you're all green. This is on the 16 port SFP+ model.
|
|
#
?
Sep 1, 2015 21:42
|
|
|
Did the mode button on the front panel get hit?
|
|
#
?
Sep 1, 2015 21:58
|
|
|
These only have UID and thankfully after years of asking we're finally putting "no setup express" in the configs of things that DO have the mode button.
|
|
#
?
Sep 1, 2015 22:36
|
|
|
What did Cisco techs do before Google? Cart around a bookshelf to reference? It seems like every day I run into an issue that is covered nowhere in any Cisco documentation and have to take hints from web searches.
|
|
#
?
Sep 2, 2015 11:49
|
|
|
Once upon a time in a land far, far away, Cisco TAC was actually good and didn't blow off your request for weeks.
|
|
#
?
Sep 2, 2015 22:10
|
|
|
CrazyLittle posted:Once upon a time in a land far, far away, Cisco TAC was actually good and didn't blow off your request for weeks. Pro-tip for TAC: have more interesting issues or make more noise to get your cases touched more. Most TAC engineers have an overwhelming amount of customers to work with and the squeaky wheel gets the grease.
|
|
#
?
Sep 3, 2015 04:22
|
|
|
DeNofa posted:Pro-tip for TAC: have more interesting issues or make more noise to get your cases touched more. Most TAC engineers have an overwhelming amount of customers to work with and the squeaky wheel gets the grease. If you can guess the solution, you get a figurative cookie.
|
|
#
?
Sep 3, 2015 04:32
|
|
|
adorai posted:it took me almost a year to get a dumbass issue resolved with tac. The symptom was if a user had two phones in two different regions with the same DN, the phones did not get the terminate signal and would keep the call "active". after 3 calls, they had to restart their phones. Use different DNs. Restart the phones after every 3 calls. Methanar fucked around with this message at 04:46 on Sep 3, 2015 |
|
#
?
Sep 3, 2015 04:34
|
|
|
Methanar posted:Use different DNs unplug the "other" phone where user isn't.
|
|
#
?
Sep 3, 2015 04:41
|
|
|
CrazyLittle posted:unplug the "other" phone where user isn't. cronjob that boots the phones at noon and midnight
|
|
#
?
Sep 3, 2015 04:43
|
|
|
Methanar posted:Restart the phones after every 3 calls. first was to upgrade the firmware. Second was to downgrade the firmware. The actual solution was to upgrade our call manager infrastructure.
|
|
#
?
Sep 3, 2015 04:56
|
|
|
adorai posted:That was the third suggestion. upgrade to "not call manager" option
|
|
#
?
Sep 3, 2015 04:59
|
|
|
That's the scrub way out. Obviously you make a new UC cluster for that user's 2nd phone.
|
|
#
?
Sep 3, 2015 05:03
|
|
|
Slightly off topic question, but I didn't see a general Enterprise Networking thread.. What's your favorite way to paint pretty pictures of your network? The stock Visio stencils are awful (at least in Visio 2010 which is what I have) and google hasn't helped me find something much better. Ideally I'd like some simple but attractive 2D shapes. (I did find Crayon Network Shapes but I think my superiors would question my sanity more than they already do if I used that.  )
|
|
#
?
Sep 3, 2015 11:53
|
|
|
Collateral Damage posted:Slightly off topic question, but I didn't see a general Enterprise Networking thread.. What's your favorite way to paint pretty pictures of your network? The stock Visio stencils are awful (at least in Visio 2010 which is what I have) and google hasn't helped me find something much better. Ideally I'd like some simple but attractive 2D shapes. Depends on what you're trying to draw. For higher-level diagrams the Cisco stencils are usually the defacto go-to. If you're diagramming complicated Layer 2/3 topologies then it's best to avoid stencils and just use simple shapes. For example, here's a snippet of a Layer 2 diagram:  If you want some fancy shiny stencils (Or physical device stencils) then checkout VisioCafe.
|
|
#
?
Sep 3, 2015 12:24
|
|
|
I just started a new job and used gliffy.com to create a network diagram for my own reference. It looks spiffy and is easy to snap together quickly, but it's awful for holding actual information. Higher ups liked it and are now tasking me with recreating the project in visio, which I'm finding impossible because visio was never meant to look good.
|
|
#
?
Sep 3, 2015 12:47
|
|
|
Obligatory network diagram 101 plug since we're on the topic of Visio and it's ilk.
|
|
#
?
Sep 3, 2015 14:29
|
|
|
ragzilla posted:Obligatory network diagram 101 plug since we're on the topic of Visio and it's ilk. This is a good read here.
|
|
#
?
Sep 3, 2015 16:50
|
|
|
With Visio you have to learn about anchor points on shapes or you are going to have a bad time when your lines go off on janky angles all the time. We use various line weights or patterns for link types, and boxes with words to describe gear. The network map isn't the place to inventory your equipment, only to be logical so you can get away with things. Make multiple 'sheets' in your VSD for extraneous.
|
|
#
?
Sep 4, 2015 12:24
|
|
|
Unity question I have a user who can receive voicemail from internal calls but not external. I've checked her settings against a known working user and they are identical. Any advice?
|
|
#
?
Sep 4, 2015 17:12
|
|
|
What does RTMT say for the call disconnect code/reason?
|
|
#
?
Sep 5, 2015 00:16
|
|
|
Hate to be that guy but is it Unity or Unity Connection?
|
|
#
?
Sep 5, 2015 00:18
|
|
|
less than three posted:What does RTMT say for the call disconnect code/reason? Didn't check yet had to move to something else. It is unity 9.x
|
|
#
?
Sep 5, 2015 01:10
|
|
|
Bigass Moth posted:Unity question You can check the standard greeting to make sure the greeting action is to take message, same with transfer. What do external callers hear? Do they hear the opening greeting? nothing? there are forwarding settings for forward unanswered/busy/etc internal.
|
|
#
?
Sep 5, 2015 03:53
|
|
|
The most likely thing is attached, for UCM anyways not CME, that all the right boxes are checked and filled in. You can hose around with this in Unity/Connection but nobody would do that without the user knowing about it. It's probably just this stupid box assuming they receive any external calls in the first place. This is the DN settings, make sure the internal/external are both checked for voicemail, or the system's pilot number is entered in the box.
|
|
#
?
Sep 5, 2015 03:57
|
|
|
Is buying a couple 5505s to study better than simulating them in GNS3? I need to keep the ball rolling in regards to practicing CLI stuff and retaining what I learned from CCNA RS while they refresh the Security exam study materials.
|
|
#
?
Sep 5, 2015 17:36
|
|
|
Speaking of ASAs, I just picked up a stack of 5510s from work and want to throw a 1GB upgrade in each. I know they're designed for ECC memory but I've read about people throwing non-ECC RAM in there. These will be strictly lab -- is there any huge issue with using non-ECC? Is the OS going to complain and fail some hardware check or should it be OK? crunk: I found that emulating ASAs in GNS3 was a huge pain in the dick. Someone told me it's better in VIRL but I wouldn't know. You may be better off getting a copy of ASAv and throwing those on an ESXi server.
|
|
#
?
Sep 5, 2015 19:51
|
|
|
Martytoof posted:is there any huge issue with using non-ECC? Is there any real cost savings? If you have sticks on hand and want to test, go for it. You won't set the thing on fire.
|
|
#
?
Sep 6, 2015 04:33
|
|
|
More Cisco UCS/Switching noob questions: What is the difference between something defined as a "fabric" vs a simple switch layout. I've noticed that our really simple UCS install (1 UCS 5108 chassis + 2 6248 fabric interconnects, we're planning on adding a second UCS 5108 chassis) doesn't have any interconnect between the A and B fabric at the interconnect level. In my mind that would mean any traffic trying to talk from one fabric to the other needs to travel all the way up to the uplink switch (which only has a total of 4Gbit) and back down the stack. Reason I ask is that we have a Nimble CS300 attached at the fabric interconnect level, and with Nimble load balancing I think both fabrics are being used by hosts when they talk to the Nimble. Also, can you link two UCS together so that they have a direct path for servers to talk between themselves and not having to go up to the fabric interconnect layer? God I wish I knew more about UCS instead of being rushed into this poo poo...
|
|
#
?
Sep 8, 2015 21:36
|
|
|
CrazyLittle posted:Is there any real cost savings? If you have sticks on hand and want to test, go for it. You won't set the thing on fire. Honestly it's just ease of procurement. I grabbed a bag of 1GB non-ECC sticks for $10 shipped. Just wanted to check whether the OS would start throwing up errors that there was no ECC present, etc. I'll find out soon though, I suppose 
|
|
#
?
Sep 9, 2015 00:29
|
|
|
Wicaeed posted:More Cisco UCS/Switching noob questions: This is correct. The two fabrics do not ever bridge anywhere within UCS and you need to use the uplink to connect the two of them together. quote:Reason I ask is that we have a Nimble CS300 attached at the fabric interconnect level, and with Nimble load balancing I think both fabrics are being used by hosts when they talk to the Nimble. I put the storage up at the uplink switch level since my NetApp also serves CIFS shares out the same ports. Putting it at the FI level would prevent that. In the Nimble case you're probably getting proper mutlipath and ALUA going on so connecting them at the FI level is probably fine. So long as you don't need anything outside the UCS environment to connect to the Nimble iSCSI you're fine and this is a pretty good setup. e: thinking about this a bit more. I'm not 100% sure on this one but: make sure that you don't have your iSCSI links in any kind of bond and have one port from each Nimble controller going to each FI instead of having both on one going to the same FI. Also make sure that you don't have fabric failover enabled on your iSCSI networks on the UCS side. When one side of the fabric is down, it should be down for iSCSI and you should be relying on multipath to do it's work. Likewise, if you have a controller die, you want to make sure that both FIs have an active path to it. You might want to check with Nimble if you didn't already. They'll know better in their case than I will. quote:Also, can you link two UCS together so that they have a direct path for servers to talk between themselves and not having to go up to the fabric interconnect layer? Nope, each server in a chassis has to at go to the FIs to talk to each other. A chassis is just a housing for server blades so the same rule applies for cross chassis communication. The FEXes on the back of the chassis are not switches and only pass traffic to and from the FIs. Connecting one chassis to another would cause errors and the ports would probably not even come up. Number19 fucked around with this message at 02:27 on Sep 9, 2015 |
|
#
?
Sep 9, 2015 02:19
|
|
|
Partycat posted:The most likely thing is attached, for UCM anyways not CME, that all the right boxes are checked and filled in. Thank you this was the problem.
|
|
#
?
Sep 9, 2015 14:36
|
|
|
I'm in week 2 of a new job moved from an Active Directory, Windows, Cisco shop to an LDAP, OsX/Linux, Juniper/Arista shop. The transition has actually been pretty smooth, and the Juniper Documentation is pretty good. Just have to put "Juniper" in the search term. Juniper Switching sucks imo, but thankfully Arista is for switching which makes it a breeze. Only thing I miss from Microsoft Office is Visio, otherwise google docs works well enough. Fun times!
|
|
#
?
Sep 9, 2015 23:34
|
|
|
Powercrazy posted:I'm in week 2 of a new job moved from an Active Directory, Windows, Cisco shop to an LDAP, OsX/Linux, Juniper/Arista shop. The transition has actually been pretty smooth, and the Juniper Documentation is pretty good. Just have to put "Juniper" in the search term. Juniper Switching sucks imo, but thankfully Arista is for switching which makes it a breeze. A very nice lightweight drawing tool can also be found at https://www.draw.io/ which I've been playing around with the past few days. I tend to find Visio to be rather cumbersome and frustrating to use at times.
|
|
#
?
Sep 10, 2015 01:06
|
|
|
My go to editor of choice for the last few years has been yEd. Cross platform and is great after you change a few default settings. It's java, but it doesn't act/perform like java. Can also output interesting formats that are html friendly if you wanted to embed somewhere, or just nice PDFs. I used Visio for the first time in years a few days ago and basic stuff link drawing a line between items and then modifying the line path afterwards seemed to be not intuitive at all. poo poo, can you even do that? Curious what you don't like about Juniper switches? Something Junos related or some broken switch feature on your platform?
|
|
#
?
Sep 10, 2015 01:22
|
|
|
Powercrazy posted:Only thing I miss from Microsoft Office is Visio, otherwise google docs works well enough. Fun times!
|
|
#
?
Sep 10, 2015 01:24
|
|
|
falz posted:Curious what you don't like about Juniper switches? Something Junos related or some broken switch feature on your platform? Interface configuration is annoying. Setting up two trunks pruning the same vlans requires a lot of configuration compared to the five lines on a range command in Arista/Cisco. Setting up two groups of ports with two overlapping switching policies is a nightmare. The multiple configuration sections for the same interface is prone to errors. I.e Ethernet-switching vs inet vs AE bundles. Bleh I just find port-grouping inflexible.
|
|
#
?
Sep 10, 2015 03:32
|
|
|
Some newer firmware removes the following from 2921 router: Ip urlfilter server vendor websense Is there a replacement command that does the same thing? The firmware I'm using is 15.4(3)m and if I can get around a rollback it would be great.
|
|
#
?
Sep 10, 2015 18:25
|
|
|
|
| # ? Apr 19, 2024 21:30 |
|
|
Powercrazy posted:Interface configuration is annoying. Setting up two trunks pruning the same vlans requires a lot of configuration compared to the five lines on a range command in Arista/Cisco. Setting up two groups of ports with two overlapping switching policies is a nightmare. The multiple configuration sections for the same interface is prone to errors. I.e Ethernet-switching vs inet vs AE bundles. I haven't worked with Arista yet, do they have something that forces you to commit or is it instant a la classic IOS?
|
|
#
?
Sep 10, 2015 22:12
|
|