Why?

From: John Aldrich 
------------------------------------------------------
Ok. Gotta ask a question that's been rattling around in my brain for 
awhile...why did 3Com/HP stop making those nifty blade switches 
(Corebuilder)??? That seems like such a great idea... you  know, buy a 
switch with room for adding more capacity, but just get what you need now, 
add extra ports as you  need them instead of having to buy a whole new 
switch.

Maybe it's just me, but that really seems smarter than buying individual 
rackmount switches.

=============================================================== From: Cameron Kilgore ------------------------------------------------------ To gouge you of your beloved operating costs. --Cameron

=============================================================== From: Bret McHone ------------------------------------------------------ Chassis based switches are still in use around. We bought two last year to replace our dated Nortel chassis based switches. They are generally meant for link aggregation rather than edge switching. -Bret

=============================================================== From: Bret McHone ------------------------------------------------------ my bad on my last reply. I misread the email... multitasking is not my best skill.. -B

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 year to replace our dated Nortel chassis based switches. switching. It really depends on your needs, and more importantly, your budget. I've = seen several data centers that use things like Cisco 4500's (and even = 6500's) as end of rack switches, and run all their edge link to that, = instead of using fixed configuration top of rack switches. Those = switches then aggregate back to beefy distribution layer switches. This is actually the way I prefer to do things, as using a chassis based = switch gives me better redundancy options than using fixed module 1u's. = If you don't need gig ports, then it's cheaper to use top of rack = switches, but if you do need gig ports, depending on the port density = needed, it's not that much more expensive to use a chassis based = switched.=20 -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.17 (Darwin) Comment: GPGTools - http://gpgtools.org iQEcBAEBAgAGBQJOBNSBAAoJEDSV5GS4KsJ4GAcH/iPCLYJRwyI+8PiYaMe5aQ8R kSehlTdHOG/JNEGPTLfTPx9ZRsMBkUqBAoBvUV1gniNtNw9g0iNweZ9oN1kaBdVL 2vEEV+YRKGsn6sWCYbxoGtlINwnsRcrKIS9O/1eyKPlelvj2mki4YjPZLfsdVjWL oyymD5a3cdrByQMZ81pS5ufFa5Y4jFI/Nd0u62eXt+T6zQ0mhl9RU3X+gcVnOCtG qT4FVY+UDk08An4bw65xCqLTlZagk+DDLB2yiMJ7mmLaceF6rQlPIvmRMWunsniJ RFi9fFVj09LdxSxxDh4w2BOwtghKJ+Y/hx+R+3Ix65io8M+9wWrZxViLRkfwHhs=3D =3DOM6T -----END PGP SIGNATURE-----

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 1. Corebuilder is kinda old, even for the 3Com Brand. 2. HP does make chassis switches, same as Cisco, Force10, etc, etc 3. This sounds like you have never purchased a chassis based switch (because) You buy a chassis based switch when uptime is paramount and port density. You can't get the same redundancy, resiliency and interconnect speed (Ok, depending on what kit you are referring to, this one is pretty much gear dependent, but) out of a "stack" of switches, or $deity-forbid, daisy-chained... Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4E120ACgkQABP1RO+tr2Rm+wCgkAYG1xXVbgaRyjaiiiaqkfQH HrUAoLub5j4OHryiI+KqXT/j8lNqX25i =SJtd -----END PGP SIGNATURE-----

=============================================================== From: John Aldrich ------------------------------------------------------ Nope, never have. I've used 'em back when I was at Covista, but never bought any. I've got a bunch of mixed-brand 1-u switches I'd like to replace with a chassis-based switch, especially if I could get more density than I currently have (probably...I've got some 12, some 24 port switches and a couple 48-port switches.)

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ok, I'll play... Budget = ? Required Feature Set = ? Nice to Have Feature Set = ? # copper ports/speed = ? # Fiber ports = ? U-Space available = ? Want Redundant Sups = ? Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4E4WYACgkQABP1RO+tr2TMVgCgow6eqT70xA1FQReUguF3vs1h mc8AoIlDxkvUfBPsH9I1qAOKOqHj9T/N =Ayq8 -----END PGP SIGNATURE-----

=============================================================== From: Aaron Welch ------------------------------------------------------ I have a 13U Catalyst 6500 that needs a new home. Any takers? -AW ty=20 =20

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Damnit Aaron.... :) Regards, dtb

=============================================================== From: Bret McHone ------------------------------------------------------ You may look at Juniper's "Virtual Chassis" when stacking switches rather than going for a chassis based system for cost effectiveness. Pick your vendor, all the big players should have some form of stacking/redundant switch. My preference leans towards brocade/foundry. Cisco just seems to enjoy pissing me off. -Bret

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 And they all suffer from the same downsides when compared to chassis based switches. Not saying they don't have their uses, and they don't have some advantages to a chassis based switch (hmm, I said plural... that might be stretching it). I haven't played w/ Brocade/Foundry in a minute, but I found F10's interesting. Interesting is another word that could be applied to Cisco as well, it sounds like you enjoy their idea of using token-ring for stacking about as much as I do.... -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4E5I8ACgkQABP1RO+tr2R8qQCdHOmcmaH6FMoYgepHwUx/92kY AggAn044wHP3cq9W0vcknGNDzMzVjgpj =WdMT -----END PGP SIGNATURE-----

=============================================================== From: Bret McHone ------------------------------------------------------ Cisco just doesn't like to play with us. They are never even in the same neighborhood in price as their competitors, let alone the same ballpark... We were able to get an 8-slot chassis and 4-slot chassis switch with awesome support for the same cost as a single switch from Cisco... they just don't want our business enough. -Bret

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Well, it's hard to compare based on these details.... but yes, Cisco can be difficultly expensive, especially for smallish enterprises. At that point, it all depends on your partner, and unfortunately, the way Cisco tiers their partners, your best bet is going to be an ATT :( And let's be honest, while Cisco is still the gold standard, there are vendors that provide more bang for your buck in the Enterprise LAN space. At least one of them still uses custom silicon too :) Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4E74UACgkQABP1RO+tr2RDdQCfaqwKw37ZE2fxMKlB/Ode7RNJ r3IAn1lsF0YXxdXblJLia1NWVicRg9Wd =7Feq -----END PGP SIGNATURE-----

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ Lemme guess, SUP1A?

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ Stackable 1U's have some issues though. Take the 3750's for example. Sure, a switch in the stack failing doesn't take the entire stack down, but it absolutely kills your backplane bandwidth. Virtual Chassis, and Cisco's VSS solution have issues too.... when those things go into split brain mode, it makes for a very long night.

=============================================================== From: Aaron welch ------------------------------------------------------ SUP2-MSFC2 Layer3 + SFC2 32-40Gbit ports. -AW

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ Yeah, we had the same issues with Cisco at my old job when we were looking to add some beefy distribution switches. Force10 was the only vendor anywhere near the ballpark of reasonable, and we ended up going with them. Once we got past the speed bump of the fact that Cisco 2950's aren't actually MST standard compliant and won't work with Force10 switches (redoing the entire networks spanning tree topology in the middle of the night = not fun), we had no issues with them. I'd like to get my hands on some Arista gear and put it through it's paces. Their business model looks interesting.

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ That's not too terribly bad then, though the space and heating issues make it unsuitable for my own use, especially if the power supplies are DC or 240 volt hehe

=============================================================== From: "Alex Smith (K4RNT)" ------------------------------------------------------ You may want to look at Allied Telesis and LG-Ericsson (formerly LG-Nortel)= . e 240

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ Yeah, I can't even consider Allied Telesis after their entire backdoor password fiasco http://www.h-online.com/security/news/item/Allied-Telesis-divulges-secret-backdoor-1251556.html?view=zoom;zoom=1 That one was a fun phone call from some freaked out execs to make sure we weren't using any of their gear. I haven't looked at any of LG's stuff. And I probably won't get the chance to either. We're cutting our support contracts with Cisco. The mothership wants us to eat our own dog food.

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I wish I could have been there.... that paints a rather amusing picture :) I'm so sorry.... Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4E95cACgkQABP1RO+tr2Rt4gCaAiamDXAOCkmNv8JXZVh3UFKA qBsAoKNzSCBrQOadJ6h+2/OIUwk0LLdZ =ZTxy -----END PGP SIGNATURE-----

=============================================================== From: John Aldrich ------------------------------------------------------ AT this point, it's just "I'd like to" replace what we have with a single chassis-based switch. I've got two different openings. I've got a 4U opening and a 6U opening. I need a minimum of 3 fiber ports (currently have two 10 mb fiber connections and a Gigabit fiber connection. I'd *like* to convert the 10mb connections to gigabit, but that'll require changing the endpoints, which is quite doable once I free up the Dell switches. *grin*) Copper ports: probably at least 75 (guesstimate) 10/100/1000 with room for more, hopefully. Redundant supervisor blades would be good so that the whole bloody thing doesn't come to a screeching halt if the supervisor blade has a problem. :D Required Feature set: Must be web manageable, with spanning-tree (or whatever the successor to that protocol is) so that when someone does something stupid like plug a switch into two two different ports, etc. it doesn't cause the network to shut down. :D I'd need to be able to do VLANs, because although I"m not currently using them, I can see a need if we actually get around to getting a SAN in the future. Nice to Have features: I really don't know what features are available or what else I would need. Several of you guys are experts on that. I'm not an expert on *anything* I'm a Jack of All Trades, Master of None. I freely admit that. :D

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'll continue to play when you bring a budget to the table :) Fiber should go into chassis as well, depending what type of fiber, how it's terminated, etc. 48 per blade is pretty common. 48 ports of GigE in a chassis will come with a price tag. 48 ports of GigE w/ wire-speed interconnects come with a big price tag. Referring back to that budget thing, supervisors are a significant portion of your investment, ensure you prepare to double up. If you need a web interface to manage a chassis, you don't need to manage it. You'll break more than you'll ever fix. Depending on load and feature set, SAN traffic is usually off-loaded to a separate (set of) switches. If we're talking about this kind of equipment and cost, you might as well climb on up to Layer 3 features. Bring a budget and I'll play architect for you. And you might want to look at your rack config again... 6U probably isn't enough for dual-sup plus at least ... say 4 blades for possible expansion? Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4E/4cACgkQABP1RO+tr2RPiwCgiaf+8hfbjIG6Q6K7Cce8rdIP uJsAmgKvHCd7m8o+gGJsMqsfaVbqBx1E =mFt1 -----END PGP SIGNATURE-----

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ If you can find a managed chassis switch that doesn't do spanning-tree, I guarantee you that you've found a company that's no longer in business ;) As far as web manageable.... *sigh* l2cli newbs! Well, that depends entirely on your needs. Generally you define the requirements for the network, and then it gets designed from there, determining what features will meet your needs.

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ Wish someone would tell that to Checkpoint.... generally, no. Most line cards are about 1RU in and of themselves, add another 5 or 6 for power supplies and general chassis form factor. You're probably not getting away with that kind of setup for less than 12u. Of course, you could just sit it on the floor and let it double as a cup holder.

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Are they still as b0rken as they were with the little triangle nokia boxes? Shhhhhh, don't give him all the answers! I could have sent a link to HP's recommended replacement for the recommended replacement for the Corebuilders (yeah, it took a bit of archive searching). Research will be good for him :) Or a copier/printer stand/table.... Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4FDvIACgkQABP1RO+tr2R4PACfWki0eZQ2oMHZTvVbE+rX9jps KzsAmwfvpbnoJbX1BCBpLRMy94ykjOro =yk5z -----END PGP SIGNATURE-----

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ Ayup. And their logs are still hellishly unhelpful. I am dutifully trying to get them exorcised from our network. I feel sorry for the poor bastards in the SOC who have to deal with our clients checkpoint firewalls on a daily basis. What are you, a chick now? ;) I don't read minds man! True, the fluid resistant chassis and linecard feature does cost extra....

=============================================================== From: John Aldrich ------------------------------------------------------ ROFL! :D Yeah... Might have to get a separate cabinet/rack just for network stuff. Or just rearrange the cabinet to consolidate space.

=============================================================== From: John Aldrich ------------------------------------------------------ Yeah..right now it was just more of a rant 'cause it doesn't seem like most VARs know what a switch like that is...

=============================================================== From: Bret McHone ------------------------------------------------------ You could look at two Brocade FCX switches. They have 2x 13Gb stacking ports with 4 1Gb fiber ports and a module slot to fit a 4port 10Gb module for uplinks as well as horizontal stacking. I have 3 of them along with Two Brocade MLX chassis switches. Plus with the Brocade support you should be able to get basic network training as well as the local engineer at your disposal.. Not trying to sell anything, just a very satisfied customer... The multi-chassis trunking is pretty slick too. -B

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Well, your original rant doesn't hold water. Chassis switches are readily available from a variety of manufacturers, in a variety of configurations, in a variety of price points. Secondly, you misunderstand the point of the chassis based switches. Yes, you figured out port density for your desire for consolidation and replacement of all the 1U switches you have, but a primary purpose of a chassis based switch is not to buy one to add as you grow. Not in the way you are thinking, because compared to the cost of supervisors, the cost of port blades are rather inexpensive. (Yes, I know there are exceptions to this, but for general purpose smallish LAN switching purposes) You're still going to have to make a large cap-ex investment to get the chassis and supervisor(s). Let me know when you do your own research if you find otherwise. And if you are phrasing the question to your VARs as a cheap initial investment to pay as you grow model for a chassis switch, then they are telling you the truth, it doesn't exist. If you're just inquiring about chassis based switches in general, you need better VARs. Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4FLPwACgkQABP1RO+tr2R7DwCePvRUbR0b9V6WHzX27ogzZpNm 4H4AnjFGxpbQAMsNTWxpRCN0c/R8tfMJ =CoE1 -----END PGP SIGNATURE-----

=============================================================== From: John Aldrich ------------------------------------------------------ I need better VARs. :D I really like chassis-based switches. It seems to make a lot more sense to me. I enjoyed using them at Covista. It's nice to get the port density in a single package rather than having to stack a bunch of switches on top of each other. Besides I like the Henry Luken approach -- buy the previous "generation" of equipment that's still good, just a bit 'out of date" and save money. :D

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ That's not entirely true. Furture growth is one of the major considerations you do make when you're investing in a good chassis switch. For example, if I know I'm going to need 10 gig ports in a couple years, I need to make sure the switch I'm buying will support the modules I need, in the numbers I need. It's a hell of alot easier to add capacity just by slotting in a new line card than it is to have to buy a new chassis, configure it, stage it, plan the downtime for it, and then actually make the swap. The decision to buy a chassis with more slots than you need in order to accomodate future growth, or to only buy one that will support what you need right now is a very common, and very important part of the purchasing process, you'd be foolish not to give the consideration it's proper weight. You're actually countering your own arguement here. The fact that it is a large capex expenditure means you do need to take future growth into consideration, otherwise if you guess wrong, you're looking at another capex expenditure down the road. This is one of the hardest things to make the folks authorizing purchases understand - the upfront cost is big, but the expansion cost per port is quite low once the initial investment is out of the way. Doing your homework in advance will actually be better on the capex :)

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ That approach works depending on your needs. If you're looking for particular features, then you have to be careful what you buy, I've seen costly mistakes made because of assumptions that given features will be present.

=============================================================== From: Aaron Welch ------------------------------------------------------ /begin rant As the person who fought every day to keep those shitty switches working eve= ry day, I should slap you. The Henry way is buying cheap shit off eBay and u= nder paying/overworking people dumb enough not to say this sucks. I have si= nce learned the virtues of support and service contracts so that I can sleep= better at night. If I interview at another company (highly unlikely) or ha= ve to do business that says support contracts are useless and too expensive,= I will promptly get up and leave. /end rant =20 =20

=============================================================== From: John Aldrich ------------------------------------------------------ *sigh* I agree with you on the maintenance contracts. I wish we could get maintenance on our ASA, but management figures it's cheaper to buy a new one than it is to pay maintenance. In theory, they're right. Paying maintenance on the ASA for a couple years will end up costing more than a new one, but it's not just the hardware costs, and I can't seem to convince management of that. As for the crappy switches, I apologize. I didn't have to do much other than help Sabados put 'em in place. They worked pretty well when I was there, except when one of Mike Cheng's guys plugged a switch into the wall twice and caused a loop. THAT is one thing I definitely want to avoid.

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ A - features like RootGuard and BPDUGuard are your friend B - That should teach you (or whomever) to know better than to leave unused ports hot.

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I should have been more specific. Yes, you will do your diligence and base your decision on your Sup/LC options, both for current needs and future needs. Unless your current needs require the absolutely most bleeding edge, fastest things you can power up (in which case, I want to come work for you ;>). But the point I was not getting to was that you're not going to buy a 6-7 (LC) slot chassis for two blades' worth of ports -- or at least John shouldn't :) No, I was addressing a different point. I was ensuring John wasn't looking for a cheap intro and build as you go with costs spread evenly, the costs will be up front. Don't you have some new (lame) switches to go play with or something? :) Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4FVRUACgkQABP1RO+tr2RxHACgsDO+wviiECwnWhp9ETJvB0re al8An3tdG3uP129fUUaxPbqpL8Ao9fc6 =wwcN -----END PGP SIGNATURE-----

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 To be fair, (or to play devil's advocate, whichever you prefer), it actually depends on your business model/needs. *Some* Networks and/or Network Segments do not require anything more than a 4 hour replaceable spare from a shelf. For the record, I would *not* classify an ISP as such a network. But I also know networks that spend their big money on core infrastructure but certain access segments don't need more than a layer 2 switch, and it's not the end of the world if it goes down for a few hours. Perhaps even worth the few grand you save over a 5 year span. Now, please, resume slapping John... Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4FVwgACgkQABP1RO+tr2RBfACgrpz80mNidYjzd+9QujPnrYW1 TO8AoLT5hATuAeNKbJfhYqIgIt0Zdo2k =c2WR -----END PGP SIGNATURE-----

=============================================================== From: Dave Brockman ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 You can get software only, and/or 24 hour response instead of the 2/4/8 hour response contracts. That should stretch the value of your 5510 further, except the re-cert fee will eat you alive.... Regards, dtb -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk4FWHAACgkQABP1RO+tr2QgTgCgjlEoeonoONgC7Tz1qFSBjmU1 lzIAoIhIolRy6meiXJ7VF+1ekCCq6sRj =Pkgt -----END PGP SIGNATURE-----

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ I would, actually. Depending on the vendor. In the case of Force10 or Juniper, the per port cost is low enough to justify having ALOT of empty slots if you forsee additional expansion in your future. You're going to get raped on the price of the iron regardless, if paying a little more up front gets me much lower costs down the road, I'm all for it. And if it gets me a bit better performance for what I do have (ie, better back plane, ability to support better supervisors, etc), then it's worth it. Ironically, in light of Aaron's rant, I'm presently stuck waiting on a call back from a vendor for support. Apparently they have better things to do on a friday night than help me fix their crap.

=============================================================== From: Bret McHone ------------------------------------------------------ iper, the per port cost is low juniper knocked themselves out of the running due to support costs for our chassis switch purchase because you have to have support on every component. Over 3years it made over $20k difference between them and brocade. ll back from a vendor for support. Apparently they =A0have better things to= do on a friday night than help me fix support should always be a consideration on what you are paying for. I will demo units to test and break to see how the support works. so far the only vendor we have really had shitty support from is JJWild who was bought by Perot Systems and then by Dell... Frisking 2 weeks waiting on them to get an esx server setup so I ended up doing it myself.... (yes I could have done it myself but politics is i

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ That's true, and a good point, I did overlook the support contract = costs, and how it's structured. It's another fair point to consider. a call back from a vendor for support. Apparently they have better = things to do on a friday night than help me fix It's funny you should mention those names, given who I work for, and = which vendor I was awaiting support from ;)

=============================================================== From: Bret McHone ------------------------------------------------------ I would just like to say that I was trying to type that last reply on an ipad, so it cut me off... I went back to my trusty old laptop. That spell check is a rant for another thread.. That vendor has been an epic failure in darn near every sense of the word. The core company of Dell is great. I have no problem with them and they helped me fix quite a few hardware issues with our tape libraries and servers promptly. This solutions group they formed from acquisition is just completely crap. They overprice everything, are slow to deliver, support is barely existent and I've had to go in behind them to fix stupid stuff... (like an SQL database placed on a C:\ drive). We were actually able to get hardware faster through a competitor in Chattanooga than that Dell Solutions group could... and THEY ARE DELL... *FUMES* Sorry for hijacking the thread... I'm the primary guy that manages all the storage, servers, networking, virtualization, wireless, etc... and my Manager helps out when he is able. This means prompt support is paramount to everything we do, so that is weighed heavily whenever we make a purchase. We try to be as self-sufficient as possible, but everyone needs helps at some point. I guess my point is, know who you are getting into bed with before you decide if the cost of the support is worth it.. I do believe support is *REALLY* important on anything you buy, but quality is what you need. If price bumps support out of the mix then I'ld suggest you look at something that is very well documented and put some cash back to hire someone on here to bail you out when you need it. -B

=============================================================== From: Kenneth Ratliff ------------------------------------------------------ -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 word. The core company of Dell is great. I have no problem with them and = they helped me fix quite a few hardware issues with our tape libraries = and servers promptly. This solutions group they formed from acquisition = is just completely crap. They overprice everything, are slow to deliver, = support is barely existent and I've had to go in behind them to fix = stupid stuff... (like an SQL database placed on a C:\ drive). We were = actually able to get hardware faster through a competitor in Chattanooga = than that Dell Solutions group could... and THEY ARE DELL...=20 Well, my employer was acquired by Dell earlier this year, and the = support I was waiting for was technically internal. However, since we're a separate business unit, we have to treat the = mothership like any other vendor. And they treat us like a client. That would be all fine and dandy, if the mothership hadn't hamstrung us = when it comes to solutions. For example, I can't even consider anything = from HP and IBM, since they're direct competitors. -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.17 (Darwin) Comment: GPGTools - http://gpgtools.org iQEcBAEBAgAGBQJOBfedAAoJEDSV5GS4KsJ4vuQH/2DM4mtGk4ZlfHcHeR2LQZX6 ZUWZpSoiM9FkuRp3MiEOiOkDWdGmKK/fx1tFrJCMBGxi1zs1MxgsOeCiOmjF7xwB 3mGlHij0jRzzQi8NkgDBrRKPDtYStmPTGmLdRwG5eUgIwGT1Nyu0W3vgxWoj5yBb TEeD7XLsRLEIP7PWD9LgVKS5Ow8MGa3qBdDOtXbVxRJQ5kJD3vDcNwirK/ojy5oh /5mW8to6O6sU4DFBMT5z3u607Ob4BADkEKMX/Y6IvIEUW+AifxLCo0dE9zDouB/c bo2F+18am62U6U1Em7bqqSgY1BjdAlcx4TQPm8l7N2v7YB+ebOZiyXDVpm4qOas=3D =3DwGux -----END PGP SIGNATURE-----

=============================================================== From: John Aldrich ------------------------------------------------------ See, I did not know that. All I knew is what our VAR offered. :(