Sorry if you were misled to believe that you were anything *BUT* cheap labor. You are. Don't feel bad, not many jobs are actually like the 'job description'. Most of the time you are doing 'other duties as assigned'.:-)
Everytime slashdot wants to push the "1000 comments" record, it posts an Evolution story. And a good measure of editorial comment to boot. (1 measure = 1 sentance).
Several other people probably got excited (as I did) when I read the headline, but too bad it excludes the 820. Slashdot has covered it before:
http://slashdot.org/askslashdot/00/12/08/2354210.s html. Hopefully it's coming soon.
I agree with Ian. Open source filtering at the firewall would be good start. And so what if it only blocks *some* or *half* (by % activity) websites, that's better than none in my opinion.
It's too bad that no one is really working hard on this.
What part of "Enterprise-ready" is not included in the 2.4 kernel?
8 or more processors? Check.
64 GB RAM? Check.
I would like to take this opportunity to point out that Windows 2000 Server only supports 4 CPUs and 4GB RAM, and even Advanced Server only supports 8 CPU's and 8GB ram. (You'd have to shell out big $$$$ for Datacenter to get the 64GB RAM support that 2.4 has).
I agree with you, 1:1 NAT doesn't really buy you any security. However, if one uses N:1 NAT (like I explained above: 2.2.2.2:port20001=>10.11.11.11), then you do get that extra security. At that point, a cracker would have to find the port first (admittedly not hard to do). But, you could make it harder by closing the port until it's time to turn on the remote access proggie (e.g. VNC).
...every single PC has to be addressable to internet...
Nope. I use VNC a lot to do remote support, and you can just punch a few holes (read: ports 21000-21050) in the firewall and NAT them through.
The other part is: call the user and say, 'click here to start the remote admin program' so i can get in first. It's a little extra protection that way.
I'm sorry, I didn't understand the question, but I might as well share what I know. Are we talking about an internet or regular network connection? If you want each person to be limited to 128Mbps, isn't that
128 * 600 = 76Gbps? That's a little more than Internet2 has to offer, isn't it?
Anyway, you could probably fine more info from:
search for the following linux programs (some free, some commercial):
ipac
iog
ipa
ipaudit
pact
bandmin
ip accounting daemon
iplog
(obviously) the 2.4.x kernel.
These projects should have mailing lists where you could ask the question of "how much hardware do i need?". I would recommend getting 1 box to do the DNS, DHCP, NAT, firewall, etc. Then have 1 box upstream throttle the bandwidth (depending on how you do your NAT, you may need to do NAT from the same box).
I'm really suprised that no one mentioned selling servers. You may *want* to sell desktops now, but trust me, once you drool over your first 8way zeon you'll never go back to desktops.
Not only is the technology MUCH cooler, but you can charge higher margins, and the competition (Dell, etc.) actually sell at a high margin as well (which gives you the opportunity to bid low). I would rather sell 1 $20,000 server a week than 20 $1,000 servers. What do you think?
Akopia, recently acquired by RedHat, developed Interchange, which is an awesome web development platform for e-commerce websites. It's GPL. I highly recommend you check it out at: http://developer.akopia.com
For many companies, the extra ten seconds it would take to spin up a backup server's hard drive(s) likely would be a non-issue.
My users get annoyed at sub-second response-time lags. What company can stand 10 seconds of wait? (Excluding NT servers, because users are used to the regular daily or weekly rebootal of their server).
Futhermore, I try to exclude every non-useful line of code from the OS and daemons that I run on the server. That includes power management.
A great resource you might utilize is the Boardwatch magazine. It's basically the equivelent of "ISP monthly". I would wager that there are some articles (or advertisements) about the information you are looking for. http://www.boardwatch.com/
Not really much of a question to go on. E.g. it makes a big difference if you are just doing a completely db-backed website verses smb sharing or something. For websites, you can setup DB replicaton now with pgsql and mysql. But the single disk system really is good, especially with GFS4.0 now out of beta.
I can't believe no one mentioned this already. I use it so often that I assumed everyone on Slashdot used it to.
2wire.com has a Free DSL Lookup Service. It takes your address and phone number, then tells you:
1. Your distance from the DSLAM (e.g. 1000 feet).
2. The Bandwidth available (e.g. 1100kbps down / 1100kbps up)
3. The Price, DSL Provider (e.g. Covad), and ISP
Secondly, I wanted to encourage you to opt for DSL instead of Cable if you can. Cable is definately the price winner, but using DSL allows you to (most of the time) chose between a large variety of competing ISP's. Whereas with Cable you have one big monopoly giving the lowest common denominator of service level (which means downtime and slow upgrade processes resulting in some months being 100kbps and other months being 900kbps).
However, if you take your time to find a good DSL ISP, you'll get one that has good server uptime, never cheats on the bandwidth, and yeilds great latencies (e.g. 20ms to yahoo.com). You should also evaluate what DSL Provider you can use. Some areas only have one avaiable (such as Covad). But where I am there are 4 choices (New Edge, Covad, NorthPoint, and Easystreet). Of those 4, when I toured New Edge's NOC, I was very impressed! They have top of the line Alcatel (#1 brand in DSL) switches and can send a packet from west coast to east coast and back in under 4 milliseconds. (THAT is good latency). I hope that helps in your search.
-Dan Browning
The company I work for specializes in out sourcing. We become your IT department. There are a lot of benefits. Imagine twenty medium sized companies that each have 1 full time sysadmin. They each pay $50,000 (made up number) yearly salary to their admin. Problems they have:
1. Vacation time. Everytime the admin takes time off, all problems get put on hold. Bad problem. And often isn't so easy to "just hire a backup admin". Unless you are talking about hiring an oursourcing company like us, then you'll be paying big bucks to have another admin on the payroll.
2. Skill sets. One admin, no matter how smart he is, will often not have ALL the tools, and ALL the skills necessary for his job. Usually this means that he spends a lot of time on the support phone learning whatever skill it is (e.g. cisco router programming) to solve a problem. Outsourcing companies like us are big enough to spread the cost of having such expertise in house. We have our own cisco router programmers that can be easily called on when their expertise is required. But this doesn't mean hiring another 50,000 employee with the expertise, it's already covered in the contract.
3. Benefits for the an admin working for us: Never has to worry about his job. It doesn't matter if a given company (customer) he is working at goes out of business or has to cut the budget or _whatever_. It doesn't affect him because he works for us. He can just go to another job site. Heck, he can go to another job site if he gets BORED. That's another benefit. Also, he doesn't have to worry about not knowing enough about any particular technology -- since he can rely on our expertise and knowledge base.
4. 24-hour support, Service Level Agreements, point-the-finger ability, etc. are some of the other problems that we can solve in our position. It's really fun to work this way. Our business mainly provides a service, but as part of that service, we can provide hardware, hosting (e-commerce or ASP), custom software developement, etc., etc. In fact, I love working here (not to mention slashdot.org browsing is encouraged).
But when those 20 companies get together and form a new company (us) that has 20 admins that are shared between the 20 companies -- it becomes much more efficient.
Specifically for this guys situation, I would say the best idea is to hire a "backup admin" through an outsourcing company like us, where they pay for someone like us to come in for whatever time he needs off.
I'm really excited to see some things starting to come about with this OSDL. Luckily, we're based out of Camas (30 minutes from the OSDL), so my company and I plan to be spending a lot of time with the OSDL (not to mention one of our employees is already enrolled in the university). So far I'm interested in the "Corporate Sponsor" option they have for $10,000, but I'm still waiting to tour the place and see what's going into it. We have a lot of open source apps and configurations that we'd like to test on some of the proposed 'big iron'. For example, I want to test out the new database clustering code that was checked into the PostgreSQL CVS a few weeks back. Maybe get a new IBM big iron in there and run a few dozen 'virtual linux's, with postgresql doing read only clustering, and then do some benchmarking. That would help us make the case for dropping some of the Oracle cluster technology for the open source solution.
http://linas.org/linux/pm.html has some good general information regarding Call Center, Bug Tracking and Project Management Tools for Linux, including open source solutions.
I have heard a lot of good reviews about KeyStone, which is free to download for personal use (including source code!) at http://www.stonekeep.com/content/download. But StoneKeep isn't open source-enlightened yet, so they still charge $$ if you use it commercially.
Slashdot Mirror
Yes, a setup.exe is available.
http://208.160.255.143/pgsql/pgsql.exe
Yes, SDL is good.
Sorry if you were misled to believe that you were anything *BUT* cheap labor. You are. Don't feel bad, not many jobs are actually like the 'job description'. Most of the time you are doing 'other duties as assigned'. :-)
Everytime slashdot wants to push the "1000 comments" record, it posts an Evolution story. And a good measure of editorial comment to boot. (1 measure = 1 sentance).
Several other people probably got excited (as I did) when I read the headline, but too bad it excludes the 820. Slashdot has covered it before: http://slashdot.org/askslashdot/00/12/08/2354210.s html. Hopefully it's coming soon.
Can word perfect write Postscript files? I know it reads/writes PDF files, so if it can do Postscript, that would be one solution.
As soon as we find a good one, please build it into Slash so we can use it for comment posting.
I agree with Ian. Open source filtering at the firewall would be good start. And so what if it only blocks *some* or *half* (by % activity) websites, that's better than none in my opinion. It's too bad that no one is really working hard on this.
I agree with you, 1:1 NAT doesn't really buy you any security. However, if one uses N:1 NAT (like I explained above: 2.2.2.2:port20001=>10.11.11.11), then you do get that extra security. At that point, a cracker would have to find the port first (admittedly not hard to do). But, you could make it harder by closing the port until it's time to turn on the remote access proggie (e.g. VNC).
...every single PC has to be addressable to internet...
Nope. I use VNC a lot to do remote support, and you can just punch a few holes (read: ports 21000-21050) in the firewall and NAT them through.
The other part is: call the user and say, 'click here to start the remote admin program' so i can get in first. It's a little extra protection that way.
I'm sorry, I didn't understand the question, but I might as well share what I know. Are we talking about an internet or regular network connection? If you want each person to be limited to 128Mbps, isn't that 128 * 600 = 76Gbps? That's a little more than Internet2 has to offer, isn't it?
Anyway, you could probably fine more info from: search for the following linux programs (some free, some commercial):
ipac
iog
ipa
ipaudit
pact
bandmin
ip accounting daemon
iplog
(obviously) the 2.4.x kernel.
These projects should have mailing lists where you could ask the question of "how much hardware do i need?". I would recommend getting 1 box to do the DNS, DHCP, NAT, firewall, etc. Then have 1 box upstream throttle the bandwidth (depending on how you do your NAT, you may need to do NAT from the same box).
I'm really suprised that no one mentioned selling servers. You may *want* to sell desktops now, but trust me, once you drool over your first 8way zeon you'll never go back to desktops.
Not only is the technology MUCH cooler, but you can charge higher margins, and the competition (Dell, etc.) actually sell at a high margin as well (which gives you the opportunity to bid low). I would rather sell 1 $20,000 server a week than 20 $1,000 servers. What do you think?
Akopia, recently acquired by RedHat, developed Interchange, which is an awesome web development platform for e-commerce websites. It's GPL. I highly recommend you check it out at: http://developer.akopia.com
For many companies, the extra ten seconds it would take to spin up a backup server's hard drive(s) likely would be a non-issue.
My users get annoyed at sub-second response-time lags. What company can stand 10 seconds of wait? (Excluding NT servers, because users are used to the regular daily or weekly rebootal of their server).
Futhermore, I try to exclude every non-useful line of code from the OS and daemons that I run on the server. That includes power management.
Nope.
A great resource you might utilize is the Boardwatch magazine. It's basically the equivelent of "ISP monthly". I would wager that there are some articles (or advertisements) about the information you are looking for. http://www.boardwatch.com/
Not really much of a question to go on. E.g. it makes a big difference if you are just doing a completely db-backed website verses smb sharing or something. For websites, you can setup DB replicaton now with pgsql and mysql. But the single disk system really is good, especially with GFS4.0 now out of beta.
2wire.com has a Free DSL Lookup Service. It takes your address and phone number, then tells you:
Secondly, I wanted to encourage you to opt for DSL instead of Cable if you can. Cable is definately the price winner, but using DSL allows you to (most of the time) chose between a large variety of competing ISP's. Whereas with Cable you have one big monopoly giving the lowest common denominator of service level (which means downtime and slow upgrade processes resulting in some months being 100kbps and other months being 900kbps).
However, if you take your time to find a good DSL ISP, you'll get one that has good server uptime, never cheats on the bandwidth, and yeilds great latencies (e.g. 20ms to yahoo.com). You should also evaluate what DSL Provider you can use. Some areas only have one avaiable (such as Covad). But where I am there are 4 choices (New Edge, Covad, NorthPoint, and Easystreet). Of those 4, when I toured New Edge's NOC, I was very impressed! They have top of the line Alcatel (#1 brand in DSL) switches and can send a packet from west coast to east coast and back in under 4 milliseconds. (THAT is good latency). I hope that helps in your search. -Dan Browning
Sorry, someone had to say it: Try upgrading them to the Samba 2_2_0 CVS branch.
I like to use mail.com because it is short and easy to understand even when your cell connection is a little fuzzy.
Octal Dump, and split the ascii output to another tty!
:-)
od $filename | tee od -a > tty2
The company I work for specializes in out sourcing. We become your IT department. There are a lot of benefits. Imagine twenty medium sized companies that each have 1 full time sysadmin. They each pay $50,000 (made up number) yearly salary to their admin. Problems they have:
1. Vacation time. Everytime the admin takes time off, all problems get put on hold. Bad problem. And often isn't so easy to "just hire a backup admin". Unless you are talking about hiring an oursourcing company like us, then you'll be paying big bucks to have another admin on the payroll.
2. Skill sets. One admin, no matter how smart he is, will often not have ALL the tools, and ALL the skills necessary for his job. Usually this means that he spends a lot of time on the support phone learning whatever skill it is (e.g. cisco router programming) to solve a problem. Outsourcing companies like us are big enough to spread the cost of having such expertise in house. We have our own cisco router programmers that can be easily called on when their expertise is required. But this doesn't mean hiring another 50,000 employee with the expertise, it's already covered in the contract.
3. Benefits for the an admin working for us: Never has to worry about his job. It doesn't matter if a given company (customer) he is working at goes out of business or has to cut the budget or _whatever_. It doesn't affect him because he works for us. He can just go to another job site. Heck, he can go to another job site if he gets BORED. That's another benefit. Also, he doesn't have to worry about not knowing enough about any particular technology -- since he can rely on our expertise and knowledge base.
4. 24-hour support, Service Level Agreements, point-the-finger ability, etc. are some of the other problems that we can solve in our position. It's really fun to work this way. Our business mainly provides a service, but as part of that service, we can provide hardware, hosting (e-commerce or ASP), custom software developement, etc., etc. In fact, I love working here (not to mention slashdot.org browsing is encouraged).
But when those 20 companies get together and form a new company (us) that has 20 admins that are shared between the 20 companies -- it becomes much more efficient.
Specifically for this guys situation, I would say the best idea is to hire a "backup admin" through an outsourcing company like us, where they pay for someone like us to come in for whatever time he needs off.
I'm really excited to see some things starting to come about with this OSDL. Luckily, we're based out of Camas (30 minutes from the OSDL), so my company and I plan to be spending a lot of time with the OSDL (not to mention one of our employees is already enrolled in the university). So far I'm interested in the "Corporate Sponsor" option they have for $10,000, but I'm still waiting to tour the place and see what's going into it. We have a lot of open source apps and configurations that we'd like to test on some of the proposed 'big iron'. For example, I want to test out the new database clustering code that was checked into the PostgreSQL CVS a few weeks back. Maybe get a new IBM big iron in there and run a few dozen 'virtual linux's, with postgresql doing read only clustering, and then do some benchmarking. That would help us make the case for dropping some of the Oracle cluster technology for the open source solution.
http://linas.org/linux/pm.html has some good general information regarding Call Center, Bug Tracking and Project Management Tools for Linux, including open source solutions.
I have heard a lot of good reviews about KeyStone, which is free to download for personal use (including source code!) at http://www.stonekeep.com/content/download. But StoneKeep isn't open source-enlightened yet, so they still charge $$ if you use it commercially.