Does anyone know if there's a way out there to implement Layer 2 VPNs on a
per-VLAN basis rather than a physical port assignment? A-La
draft-kawakami-mpls-lsp-vlan-00.txt.
Ideas welcome,
Karen Y
A rose by any other name is Cisco specific terminology...
Message Posted at:
***
On 7/15/2003 at 7:29 AM Zsombor Papp wrote:
At 09:48 AM 7/15/2003 +, Karen E Young wrote:
KY: According to the RFC (page 99) If the Interface MTU field in the
Database Description packet indicates an IP datagram size that is larger
than the router can accept on the receiving
/2003 +, Karen E Young wrote:
KY: According to the RFC (page 99) If the Interface MTU field in the
Database Description packet indicates an IP datagram size that is larger
than the router can accept on the receiving interface without
fragmentation,
the Database Description packet is rejected
Comments inline...
*** REPLY SEPARATOR ***
On 7/12/2003 at 10:15 PM Hemingway wrote:
Zsombor Papp wrote in message
news:[EMAIL PROTECTED]
At 07:54 AM 7/12/2003 +, Hemingway wrote:
hebn wrote in message
news:[EMAIL PROTECTED]
layer 2 frame has a MTU of 1500
Simon,
Careful there! Don't mistake MLS for MPLS. Two different animals entirely.
MLS is Multi-Layer Switching and is strictly a LAN technology while MPLS
(Multi-Protocol Label Switching) is predominantly used in the WAN.
Here's some stuff on MPLS.
John,
Your best bet is to take either the composite exam or the recert exam. Both
will get you recertified and both are only one exam rather than 4. If you
can wait until the composite exam comes online (August 7th) that one may be
best since it looks like it will recertify both CCNP and CCDP at
If you check Cisco's website, you'll see that the old Foundation exam and
both the CCNP recert and the CCDP recert all go offline on September 7th.
The new test (642-891) will replace them and comes online on August 7th.
The current CCNP and CCDP recerts will no longer be offered after September
Think about it...
With Hybrid mode, the CPUs on the MSFC and the supervisor engine function
independently. You have two user interfaces. With Native mode, the two
devices are integrated with a single user interface and the CPU on the MSFC
functions as the primary CPU for the switch.
Karen
A couple weeks ago I put out an email with links to all of the general
sessions from 1998 to 2002 as well as links to the power sessions from 2000
to 2002. The message subject line read FYI: Cisco Networkers Links [7:64652]
Karen
*** REPLY SEPARATOR ***
On 3/22/2003 at 7:58 PM
Eric,
The gratuitous ARP is just to let the switch or bridge know that the port
that the virtual MAC is attached to has changed.
If an existing router is converted to HSRP, then the end stations will
continue to track the real MAC address, not the virtual one. You have to
reboot the end stations
Your router is connected to a port on a frame-relay switch on the provider's
end. That port can handle multiple PVCs. a DLCI is just the FR switch's way
of determining which of those PVC's a frame belongs to and thus how it
should be routed. So from a certain viewpoint, both statements are true.
you
heard of that?
The workaround is to use the trunk or nonegotiate keywords to force
interfaces in different domains to trunk.
Thanks,
Priscilla
Karen E Young wrote:
As long as the native VLAN is the same on both ends so that the
ends of the prospective trunk link can communicate, DTP
My take on it is... its out-of-band. It's just multiple logical channels
multiplexed onto a single physical channel. It doesn't matter that the
logical channels work together, the time slots remain dedicated to their
respective channels and the traffic doesn't mix.
Just my .02
Karen
***
Converge Network Digest has a few tutorials on MPLS on their web site. Not
alot of detail but really good overviews for someone new to the technology.
http://www.convergedigest.com/Bandwidth/archive/010910TUTORIAL-rgallaher1.htm
Hope this helps,
Karen
*** REPLY SEPARATOR ***
Etherchannel is a way of bundling together multiple links between switches
or between a switch and a router so that they function as a whole.
Here's a link to a white paper about it on Cisco's site.
http://www.cisco.com/en/US/tech/tk389/tk213/technologies_white_paper09186a0080092944.shtml
Hope
They could. In fact, its quite likely.
The link from your CPE goes into a port on one of their WAN switches. from
there it goes over a trunk utilizing either Fast Packet (FP) or ATM to
another WAN switch. There may be a number of WAN switches between your CPE
and the destination CPE. You can get
As long as the native VLAN is the same on both ends so that the ends of the
prospective trunk link can communicate, DTP will be able to form the trunk.
The VTP domain is irrelevant. All DTP needs is layer 2 connectivity and the
desire (on both ends) to trunk. :-)
In fact, one of the requirements
Comments in-line
*** REPLY SEPARATOR ***
On 3/10/2003 at 4:03 AM Steiven Poh-\(Jaring MailBox\) wrote:
Hey Guys,
Can anyone please help answer just few question for my assigment? Thanks
1) what is the difference of ethernet 2 and ethernet SNAP frames ?
Cisco's Ethernet
Greetings!
I've recently gathered up all my links for Cisco Networkers and it amounts
to a fair amount of stuff. I just thought that everyone should have the
benefit of this stuff so here it is. There's some good stuff on Network
Design in the 1999 General Sessions. Look at the bottom half of
Since ATM wasn't an IETF standard, RFCs aren't the best choice for
information.
Here are some good references:
DISA - Commercial Communications Standards - Approved ATM Forum
Specifications:
http://www-comm.itsi.disa.mil/atmf/index.html
ATM Forum:
http://www.atmforum.com/
IEC (International
Here -
http://people.ee.ethz.ch/~oetiker/webtools/mrtg/
Almost everything you need to know can be found there.
The comp.dcom.net-management newsgroup is a good forum for getting it set up
the way you want it. Both MRTG and RRD Tool (comes with MRTG).
If you want to run it on a Windows server,
James,
On a switch, the VLAN that the IP address is assigned to defines the
management interface. The switch can only have one management interface and
thus, only one IP address. When the IP address was assigned to VLAN 1 (the
default) then devices that connect directly to ports attached to VLAN
Rutger,
Cisco s2witches allow you tyo specify which VLANS will be allowed to
traverse a trunk link. Your first example simply identifies the port as a
trunk link without any limitations as to which VLANs can use it. Your second
example sets up the port as a trunk link but limits the traffic to
A native VLAN is the VLAN that that port uses when trunking breaks down.
Thats it. If you don't set it to a specific VLAN in the config, then the
native VLAN will be the default vlan. On cisco, this is VLAN 1.
Normally, the trunk is up and running and the native vlan doesn't come into
play.
Tunde,
It sounds like your users are getting blocked by spanning tree on bootup.
Since the switch is spending ~50 seconds running spanning-tree before it
forwards any data, the DHCP requests aren't even getting through.
PortFast bypasses the usual spanning tree thing. It allows the port to
switchport mode trunk
spanning-tree portfast
From: Karen E Young
Reply-To: [EMAIL PROTECTED]
To: Emile Harding
CC: [EMAIL PROTECTED]
Subject: Re: Cisco VLAN Help-Group Study [7:62293]
Date: Fri, 31 Jan 2003 18:45:00 -0800
MIME-Version: 1.0
Received: from mclean.mail.mindspring.net
Emile,
Here's what I see right off hand...
1) You aren't trunking. The switch isn't set up for it.
Pick a port to connect the switch to the router with and configure it to
trunk. Make sure that it isn't set up with a VLAN as this can interfer witht
eh trunking.
Example, if you want FE0/1 to be
Since the interfaces are directly connected it shouldn't matter. The routes
are already there. Doesn't need EIGRP for that. Remember your administrative
distances.
*** REPLY SEPARATOR ***
On 2/1/2003 at 2:02 AM Daniel Cotts wrote:
The router has under router eigrp 100 a
Now, now... the OSI model DOES have a purpose. It is a reference model - the
key word being reference.
It isn't the model so much as the fact that it makes it alot easier to look
at the various processes discretely rather than as a whole. The OSI model
gets you used to the idea that a process is
Elmer,
Since an IP address needs to be mapped to a MAC address for delivery, a
multicast frame needs a destination MAC address in the header. As a
multicast frame is going to multiple destinations that are probably not
known to the sender, a special MAC address needs to be used. After all, you
Message-
From: Karen E Young [mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 13, 2001 2:00 PM
To: [EMAIL PROTECTED]
Subject: RE: about multicast address! [7:29057]
Elmer,
Since an IP address needs to be mapped to a MAC address for delivery, a
multicast frame needs a destination MAC address
the device
belongs to.
Just shows what happens when you try to do too many things at once
Karen
*** BEGIN FORWARDED MESSAGE ***
On 12/13/2001 at 3:27 PM Karen E Young wrote:
Elmer,
In fact I have done soem teaching, however, it was the months spent doing
phone-tech-support
Hmmm...
A layer-3 protocol (BGP) encapsulated in a layer-3 protocol (IP). Does this
sound anything like tunneling to anyone else? :-)
Sorry, slightly manic at the moment...
Karen
*** REPLY SEPARATOR ***
On 12/13/2001 at 4:41 PM Howard C. Berkowitz wrote:
Just an open
Sergio,
Here's a link to Cisco's internal IP Multicast Training.Thought it might be
of use to you.
ftp://ftpeng.cisco.com/ipmulticast/training/index.html
As for mapping addresses...
ftp://ftpeng.cisco.com/ipmulticast/whitepapers/technology_overview/index.html
Hope this helps
*** REPLY
I haven't personally bought any, but a friend of mine has and seems pretty
pleased. They're just recordings of the Networkers sessions, so each is
roughly 2-4 hours long.
Personally, I would suggest the CDs rather than individual session tapes.
They're in MP3 format, but you get a discount. Per
I found that if I take a break from studying, it can be really hard to get
back into it so I came at it from a different direction. I realized that it
wasn't the studying that I was burnt out on, it was the monotony. The same
subject all day every day. I decided that if I was burnt out on
SNMP just polls the interfaces listed in ifIndex. It doesn't care whether
those interfaces are virtual or not. Virtual interfaces get a listing in
ifIndex just as physical interfaces do. Thats why the ifIndex values can
change if you change the config of your router or switch. Adding and
removing
Kell is right about things being a mess later on.
The default vlan is VLAN 1, which carries the protocol control traffic for a
switched network. Things like STP BPDU's, DTP (ISL and dot1q), VTP, PAgP,
etc... In addition, if the management interfaces are set to VLAN 1 (the
default), then
If you're getting FECN/BECN traffic then its experiencing congestion.
Congestion on the FR network prior to CIR being met is indeed a good sign of
oversubscription. If its only occasional then it may simply have been a
spike on the switch. However, if you're getting it fairly regularly, then
you
Here's a link for stuff that I used when I took the exam. It covered all of
the stuff I ran into on the exam.
http://cisco.com/univercd/cc/td/doc/product/wanbu/8_5/switch/sys/sysmpt01.htm
As an alternative, Chapter 9 of Cisco WAN Quick Start (ISBN 157870104X)
summarizes all of that info in a
Here's some info.
http://www.protocols.com/pbook/ibm.htm
http://www.redbooks.ibm.com/pubs/pdfs/redbooks/sg242009.pdf (Chapter 4)
HTH,
Karen
*** REPLY SEPARATOR ***
On 8/27/2001 at 5:35 AM Lists Wizard wrote:
Hello Group,
I tried my best to find information about
Dave,
You can do some filtering. There are specific access lists that you can use
for SAP filtering (access-list 1000-1099) and for filtering NLSP route
aggregation (access-list 1200-1299).
http://www.cisco.com/univercd/cc/td/doc/product/software/ios112/112cg_cr/6rbook/6ripx.htm
HTH,
Keep in mind that PAgP (Port Aggregation Protocol) is the protocol that sets
up and tears down EtherChannels. Unless the port is configured to shut off
PAgP then you'll still get activity for it since the default setting is
auto. Even on a port that ISN'T configured for EtherChannel. You might
Hamid,
On a switch, the only interface that requires an IP address is the
management interface. The management interface is a virtual interface that
isn't attached to a physical port and is of use only for management - hence
the name. It handles such management protocols as telnet, TFTP, and
Priscilla,
A native VLAN is the VLAN that a port functions under when its not trunking.
When the switch comes online one of the things it does is perform any
negotiations for trunking. In order to negotiate the two ports need to be
able to communicate at layer 2. If the ports at either end of
Kwame,
Make sure you have the correct cable type (normal or crossover). Otherwise,
try swapping it out for another that you know is good. Given your config,
its probably not a duplex issue. You should do a show int on the affected
ports to double check that the ports are actually functioning in
Another thing you might try is your local library. If your local library
doesn't have it, you can try an inter-library loan. Just find the ISBN
number of the book(s) you want and put in a book order. ISBN numbers are a
bit more dependable as an identifier than the title or author.
HTH,
I did some research on this last summer. At that time 2900Xl and 3500XL
switches did not support RADIUS authentication. (Switch IOS version
12.0(5)XP). As it happens, they didn't support Kerberos Telnet either but
you didn't ask for that.
Here's a quick sampling on what I got:
Cat4000 ver
CM,
The name of this protocol is misleading (VTP = VLAN Trunking Protocol) since
it doesn't really have anything to do with trunking. VTP is merely a way to
update the VLANs that a switch recognizes as valid on multiple switches by
updating a single server. All of the rest of the switches in the
Ralph,
I've gotten this message before myself. It occurred when the two ports that
made up a trunk link had different native VLANs. Changing the native VLAN of
one of the ports to match the other allowed the trunk link to come up. I
Imagine that your situation is likely quite similar. just check
Michael,
I'm no specialist, but aren't mapped drives tied in with NetBEUI? Is your
WAN link set up to carry this traffic?
Just a thought...
Karen
*** REPLY SEPARATOR ***
On 6/29/2001 at 4:02 PM Michael Montgomery wrote:
Does anyone know if there are any possible
I did some research on this about a year ago for a project at work.
As of version 5.4(3) of the catalyst OS code, local Usernames and Passwords
were not supported. Sorry guys, you can't do it.
HTH,
Karen
*** REPLY SEPARATOR ***
On 6/30/2001 at 4:56 AM Gareth Hinton
Hamid,
You could always try an access list that filters ethernet addresses (list
numbers 700-799). Its not commonly used so people seem to forget about it.
HTH,
Karen
*** REPLY SEPARATOR ***
On 6/30/2001 at 8:41 AM Hamid wrote:
Hello group
How can I deny a couple of
I have to agree with the subnetting thing. IP addresses are onlywri
*** REPLY SEPARATOR ***
On 6/29/2001 at 8:24 PM Howard C. Berkowitz wrote:
Open Forum might be more up to speed if your free-time consists of working
address plans in binary or looking for probability in the
Art,
Given the nature of the data, you could graph it as a gauge integer(Options:
gauge) but the graph really wouldn't give you any practical data (boring to
look at and easy to miss the blips). Besides, it wouldn't give you uptime -
just the average of the values gathered. In broad terms,
or wrong? I know
how to make it work either way, but I'd just like to get an idea of what
the better ways are of separating Internet bound traffic from LAN based
traffic?
Vijay Ramcharan
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of
Karen E You
Here's an excerpt from the RFC 1213 MIB with the descriptions for
ifOutUcastPkts and ifOutNUcastPkts... The same descriptions work for
ifInUcastPkts and ifInNUcastPkts as well. Your assumption that the total
packets out is thesum of ifOutUcastPkts and ifOutNUcastPkts is correct.
ifOutUcastPkts
John,
You really want to limit www traffic don't you? First you allow traffic _to_
203.111.42.204 or 203.111.42.215. If the www traffic isn't caught by that
then it needs to match a source of 203.111.42.224, 203.111.42.225, or
203.111.42.226. Everything else is going to get caught by that
Vijay,
All you need is a default gateway on the router that points to the internet.
When an Internet destined packet from a workstation on a VLAN hits the
switch it gets dumped off on the router or MSFC since it doesn't have a
destination MAC address of a device on that VLAN. The router takes
Greetings,
Question 1) Yes - using the ip helper address command on the router
interfaces for the VLANs that don't have a DHCP server.
Question 2) Technically, yes you can. But only by bridging on the router
interfaces for the VLANs. However, it does fall under the first rule of
networking,
Remember that in a switched environment the issues change from a
non-switched network. The problem with collisions almost disappears and the
issue becomes broadcasts. When an end-node receives a broadcast it shuttles
the packet up the stack and the NIC issues an interrupt. The problem with
large
Comments inline...
Let me know if you have any questions.
HTH,
Karen
*** REPLY SEPARATOR ***
On 6/12/2001 at 5:34 AM Amit Gupta wrote:
Hi All,
The scenario is such that I have 2 subnets configured
on the LAN. They are x.x.1.0 / 24 and x.x.2.0 / 24.
The IP address for
Albert,
I had a case where one of the VLANs on the internal network should only
access the internet, not the internal portion of the network. The purpose
was to allow guests at the company to hook in and access the internet
without giving them access to the internal network or using a modem and
Here are some resources
RFC 1195
RFC 2763
Configuring Integrated IS-IS
http://www.cisco.com/univercd/cc/td/doc/product/software/ios120/12cgcr/np1_c/1cprt1/1cisis.htm
Tutorial: Introduction to IS-IS
http://www.nanog.org/mtg-0010/isis.html
OSPF and IS-IS - A Comparative Anatomy
And where that leaves off you can try this.
Optical Networking Crash Course ISBN - 0071372083
Karen
*** REPLY SEPARATOR ***
On 6/7/2001 at 5:29 AM Dom Stocqueler wrote:
Try the following site -
http://www.lightreading.com/
it has lots of stuff from beginner's
Iyuri,
You need a 100Mb ethernet port to do trunking, which the 2501 doesn't have.
A 2600 series will though. If you don't use trunking, then you will need a
10Mb ethernet port on the router for each VLAN you want to route between. In
this instance you're just setting up standard access links
Its not Cisco but
Heres the user manual for version 3.2.0.0 of the software for that router.
http://www.enterasys.com/support/manuals/hardware/2578_07.pdf
The section you're looking for is in Chapter 9 P.99 (Acrobat reader shows
page 125 of 270)
ssr(config)# ip helper-address
The requirements for connecting a router to a switch running multiple VLANs
have been posted to the list before do I won't go into great detail.
If VLAN 1 has an IP address on the switch then you need to set up an IP
address for whichever VLAN you want to act as the management VLAN and THEN
shut
but the switches were pretty picky
about it. This is where setting up VLAN 1 as the management interfaces/VLAN
came in handy most. Since the default VLAN of a port is VLAN 1, I didn't
have to change it.
HTH,
Karen
*** REPLY SEPARATOR ***
On 6/6/2001 at 2:49 PM Karen E
Ronnie,
1. If you mean PVCs in general rather than Cisco in particular then the two
phases (I'd say phase is a better term here and less likely to cause
confusion) you're asking about are transfer and idle. Since there isn't any
call setup and termination phase for the link that leaves only the
Arun,
I'm not completely sure, but I think that you have to run 12.0 or higher to
do trunking on a router. You might want to double check that though.In the
mean time, here are examples of a working trunking config on a 2924XL that I
work with and the complementing router config.
Carmelo,
You might want to double check trunking support on that cat 4000. Some
models do not support ISL and only have Dot1Q support.
Regardless, it would be better to only have one trunking type on the
network, if only for troubleshooting purposes.
Hope this helps,
Karen
***
Andrew,
I don't have a link right off hand but I can give you the main reasons why
its a bad idea. You don't want to mix user traffic with control and
management traffic.
Certain protocols always run over VLAN 1. These protocols include VTP, STP,
CDP, PAgP, and DISL/DTP. Other protocols use the
John,
Go to www.groupstudy.com. On the right hand side about half way down there's
a List Manager heading. It allows you to subscribe to the digest version
of the professional list.
HTH,
Karen
*** REPLY SEPARATOR ***
On 5/4/2001 at 2:49 PM John O. Watkins wrote:
Does
You might want to look at the UplinkFast and BackboneFast features. These
features allow fast switchover to redundant links on switches running
spanning-tree in the case of a failure.
HTH,
Karen
*** REPLY SEPARATOR ***
On 5/4/2001 at 11:54 AM Kim Seng wrote:
I have a
Sammi,
At my last position I was called in to redesign a smallish corporate
network. They had 23 VLANs running on a 2621 router. The average CPU
utilization was in the range of 30% to 40%. When I dropped it to 4 VLANs the
utilization dropped to between 1% and 3%. While I'm sure that the 3600
I stand duly correct by both of you. :-)
Please forgive the SNA goof. Its not something that I've ever worked with so
it doesn't stick in my mind all that well.
As for the HDLC/LAPB... I know better than that, I just wasn't thinking. All
I can say in my defense was that I had just gotten home
Sammi,
You can always create VLANs without a router, you just can't move traffic
between them. If you have a situation where you have two networks that exist
in the same location but need to be kept strictly seperate (such as a
production and a test network) then it isn't necessarily a bad thing
Kuldip,
For most of these, a quick look at the purpose of the protocol will answer
your question. However, don't assume that error control is required,
sometimes there isn't any error control at all. However, even when a
protocol has error control functionality, don't assume that it will do
Jake,
Technically, not all Cisco switches are Catalysts, only the LAN switches, WAN switches
aren't (well...they ARE switches..)
Anyway:
IOS:2900XL, 3500XL, 4840G, 6000
Cat OS: 2900G, 4000, 5000, 6000, 8500
Menus: 1900, 2820, 3000 (3100, 3200), 3900 Token Ring
Does this sum things up
When you connect to a brand new router for the first time you need to use a console
connection because there isn't an IP address yet to allow you to connect via telnet.
Its the same thing with switches and hubs. If you don't have an IP address on the box,
you're reduced to using console
MPLS Tutorial:
http://www.nanog.org/mtg-9905/ppt/mpls/index.htm
MPLS Resource Center:
http://www.mplsrc.com/
Internet Drafts on MPLS:
http://www.ietf.org/ids.by.wg/mpls.html
Cisco info on MPLS (a start anyway):
Kiran,
It depends on the protocol that you run over the serial interface.
By default, synchronous serial interfaces operate in full-duplex mode . However, the
usage of certain protocols changes that default to half-duplex.
More info here:
Mark,
You can use a local security database (Username...) to set up user accounts with the
privilege levels you want.
Here's some URLs that have info and config samples.
http://www.cisco.com/univercd/cc/td/doc/product/software/ios113ed/113ed_cr/secur_c/scprt5/scpasswd.htm
show interface port-channel
*** REPLY SEPARATOR ***
On 2/14/2001 at 8:31 AM Pierre-Alex wrote:
:Hi Ahmed,
:
:I do not have the command show port channel on the 1912 Switch.
:Do you know another way of checking this out?
:
:Regards,
:
:C1912#show port ?
: blockForwarding
Probe the ports to see if they're responding? Use a Perl script to verify that the
service is running? Use the "dhcpcmd" utility from the resource kit ("dhcpcmd dhcp
server IP mibcounts") to grab the stats?
Just a few thoughts...
K.Young
*** REPLY SEPARATOR ***
On 12/4/2000
Saw a preview of this one at Networkers in vegas. Looked like a good book to have. I
assume that the final version would be evenb better since the version I saw was kind
of ... unfinished :-).
Karen
*** REPLY SEPARATOR ***
On 8/15/2000 at 11:59 PM Andy wrote:
just got a new
rimary may be router A for VLANs 1-5 and Router B for VLAN 6-10. It all depends on
how you set it up. Of course, the same could be said even if you aren't using HSRP.
Although I'm not sure why you wouldn't when you clearly set up rthe network with
redundancy at the switching core (the 6500s).
Karen E Yo
Funny, I thought that bread was an encapsulation method and was strictly determined by
the payload. For example, Encapsulating corned beef and saurkraut in white bread WILL
cause transport failures and a corruption of the mustard whereas the more stout "Rye"
encapsulation method was designed
Ronald,
Strictly speaking, the Cat 5000 series doesn't use IOS. It uses the Catalyst Switching
software. Slightly different beast. The newer switches are coming outr with the
switching version of the IOS instead so it looks like the old Catalyst software is on
the way out. It's kind of a pain
No, you don't need a router. A node is determined to be a member of a VLAN by their
MAC address. Layer 2 rather than Layer 3 remember? The router is only needed to deal
with packets destined for anything outside the VLAN. The IP addresses don't matter.
Karen E Young
*** REPLY
The "Customer Appreciation Event" was much more fun! The trolli hat was a big hit when
I got back to work. :-)
*** REPLY SEPARATOR ***
On 6/27/2000 at 7:46 PM Phillip Diamanti wrote:
:Beer bash at networkers who
. Lucky I guess.
Karen E Young
Network Engineer
ELF Technologies, Inc.
*** REPLY SEPARATOR ***
On 6/9/2000 at 2:52 PM Andrew Larkins wrote:
:Carry on studying. I have been in the industry for a while now, and have yet
:to see a lady engineer. I would actually like to see more
93 matches
Mail list logo
