(TIL) Today I Learned...

Started by Seittit, January 13, 2015, 03:50:21 AM

Previous topic - Next topic
Print
Go Down Pages 1 ... 9 10 11 12 13 ... 34
User actions

Otanx

#150
July 08, 2015, 10:45:36 PM
I am learning a lot this week...

TIL that Oracle really needs to stick to databases, and away from networking. A quote from Oracle documentation on networking configurations for Oracle RAC deployment.

"STP events should be contained, either by disabling STP for the VLAN or implementing vendor
specific STP re-convergence optimizations such as Port Fast definitions, RootGuard, BPDU
filtering."

When a database guy reads this his eyes glaze over, and he says "Oh, disable spanning-tree. Oracle says you should." Then I get to read Oracle documentation to figure out what he is talking about.

-Otanx

icecream-guy

#151
July 09, 2015, 07:09:39 AM
Quote from: deanwebb on July 08, 2015, 07:56:12 PM
TIL that when AD is in a mixed 2003 and 2012 environment, everyone blames NAC when people get kicked off the wireless... because of AD doing weird stuff with reading the user certificates on the mobile devices...

TIL, you got like 5 days to get off that EOL 2003 OS....  reminds me. I got to rebuild my Dell 1300 Windows 2003 server into something else, but that's another thread.
:professorcat:

My Moral Fibers have been cut.

hizzo3

#152
July 13, 2015, 10:07:10 PM


Quote from: Otanx on July 08, 2015, 10:45:36 PM
A quote from Oracle documentation on networking configurations for Oracle RAC deployment.

"STP events should be contained, either by disabling STP for the VLAN or implementing vendor
specific STP re-convergence optimizations such as Port Fast definitions, RootGuard, BPDU
filtering."

When a database guy reads this his eyes glaze over, and he says "Oh, disable spanning-tree. Oracle says you should." Then I get to read Oracle documentation to figure out what he is talking about.

-Otanx

So when do we get to sit back and watch the network go down like a city at night after a substation trips the entire regional grid? :)

FilipiNomad

#153
July 17, 2015, 08:47:12 PM
TIL that, by default, all downstream ports on the 3400 are UNI ports and pings are dropped on UNI ports.  I had a PC connected to a downstream port.  I set up an SVI and tried pinging that but failed.  Disabled the firewall on the PC and that didn't work.  Tried a different cable, port, PC and it still didn't work.    Checked the ARP table on the PC and I can see the switch.  Checked the ARP table on the switch and I can see the PC.
:jackie-chan:

Looked up the configuration guide for the switch and found that the switch has a Control Plane Security feature that drops ping on UNI ports by default. :doh:

deanwebb

#154
July 17, 2015, 09:27:46 PM
Oh man, you're going to look back on that and laugh one day.
Take a baseball bat and trash all the routers, shout out "IT'S A NETWORK PROBLEM NOW, SUCKERS!" and then peel out of the parking lot in your Ferrari.
"The world could perish if people only worked on things that were easy to handle." -- Vladimir Savchenko
Вопросы есть? Вопросов нет! | BCEB: Belkin Certified Expert Baffler | "Plan B is Plan A with an element of panic." -- John Clarke
Accounting is architecture, remember that!
Air gaps are high-latency Internet connections.

wintermute000

#155
July 18, 2015, 03:29:30 AM
Today I learnt about this charming Field Notice. Three guesses how.

http://www.cisco.com/c/en/us/support/docs/field-notices/637/fn63743.html

Problem Symptoms

If the suspected Catalyst 6500 supervisor, linecard, or fixed configuration hardware has been in operation for approximately 24 months, the product hardware might fail to boot up due to memory failure during a power cycle event. This is caused by one or more of these actions:
Upgrade the software
Reload the entire product
Reload after installation
Chassis power cycle
Online Insertion Removal/Replacement (OIR)
Note: This issue does not affect boards while the boards are in operation.  The board failure might occur after one or more of the actions listed are executed.


wintermute000

#156
July 18, 2015, 03:30:21 AM
Quote from: FilipiNomad on July 17, 2015, 08:47:12 PM
TIL that, by default, all downstream ports on the 3400 are UNI ports and pings are dropped on UNI ports.  I had a PC connected to a downstream port.  I set up an SVI and tried pinging that but failed.  Disabled the firewall on the PC and that didn't work.  Tried a different cable, port, PC and it still didn't work.    Checked the ARP table on the PC and I can see the switch.  Checked the ARP table on the switch and I can see the PC.
:jackie-chan:

Looked up the configuration guide for the switch and found that the switch has a Control Plane Security feature that drops ping on UNI ports by default. :doh:

This is where the RTFM comes in handy. Metro ethernet switches are not normal switches, they are much funkier.

deanwebb

#157
July 18, 2015, 10:13:45 AM
:rtfm:

New Smiley. The code should be obvious.
Take a baseball bat and trash all the routers, shout out "IT'S A NETWORK PROBLEM NOW, SUCKERS!" and then peel out of the parking lot in your Ferrari.
"The world could perish if people only worked on things that were easy to handle." -- Vladimir Savchenko
Вопросы есть? Вопросов нет! | BCEB: Belkin Certified Expert Baffler | "Plan B is Plan A with an element of panic." -- John Clarke
Accounting is architecture, remember that!
Air gaps are high-latency Internet connections.

NetworkGroover

#158
July 18, 2015, 12:13:47 PM
Quote from: Otanx on July 08, 2015, 10:45:36 PM
I am learning a lot this week...

TIL that Oracle really needs to stick to databases, and away from networking. A quote from Oracle documentation on networking configurations for Oracle RAC deployment.

"STP events should be contained, either by disabling STP for the VLAN or implementing vendor
specific STP re-convergence optimizations such as Port Fast definitions, RootGuard, BPDU
filtering."

When a database guy reads this his eyes glaze over, and he says "Oh, disable spanning-tree. Oracle says you should." Then I get to read Oracle documentation to figure out what he is talking about.

-Otanx

Well duh - no spanning tree - no convergence issues!  Win!
Engineer by day, DJ by night, family first always

FilipiNomad

#159
July 18, 2015, 10:09:46 PM
Quote from: wintermute000 on July 18, 2015, 03:30:21 AMThis is where the RTFM comes in handy. Metro ethernet switches are not normal switches, they are much funkier.

I've done something similar to a Ciena ME switch and it worked fine, so I didn't think I'd run into any issues.  I was actually just helping somebody configure their 3400 and told them it should just take a few minutes.  One hour later... :wall:

wintermute000

#160
July 18, 2015, 10:47:08 PM
You'll find that UNI vs NNI is quite common in SP metro ethernet. Its all over the JNCIS-SP material for example - these are Metro Ethernet Foundation standards, not cisco

FilipiNomad

#161
July 19, 2015, 09:41:50 AM Last Edit: July 19, 2015, 10:08:41 AM by FilipiNomad
I'm familiar with UNI vs NNI. Almost all my projects are MetroE and MetroE-over-DWDM using Ciena gear. I've configured a Ciena MetroE switch similarly and it worked fine. The Ciena behaves differently. I didn't realize that Cisco downstream ports were UNI by default, and I haven't  read anything on the MEF-CECP study materials about pings being blocked on UNI ports.

Nerm

#162
July 21, 2015, 12:04:13 PM
TIL that if you have an existing backup config password recoveries on Watchguard XTM's are pretty easy.  :cheers:

wintermute000

#163
August 08, 2015, 05:15:48 AM
TIL that if you generate a CA cert then sign a cert for the same box, do NOT throw the IP address of the box into the CN on both certs.... i keep telling work I'm not a security guy :p

routerdork

#164
August 11, 2015, 03:29:02 PM
TIL how to use Wireshark on the Cisco 3850's. Pretty slick.
"The thing about quotes on the internet is that you cannot confirm their validity." -Abraham Lincoln
Print
Go Up Pages 1 ... 9 10 11 12 13 ... 34
User actions