Can't everyone just get back to what they're best at: making truckloads of cash?
I mean, Intel beat AMD in the desktop micro space, and NVIDIA beat ATI in the desktop GPU market. The two losers even had to join forces to stay afloat.
Seems to me that everyone would be happier if they just played nice, and counted their money.
However, I'd think that load balancing those connections would have been the primary issue. Keeping a stateful IMAP connection open across a given cluster of servers can be tricky, versus a straight "wham, bam, thank you ma'am" POP login.
This would generally be a problem only for mere mortals, but Google is, well... Google. So I'm not surprised to see them offering it, even if it is a little late in coming.
I've worked in many different environments, both as a staff member and as an outside consultant.
After the "real" IT staff, who do I go chat with about a given system's issues? The secretaries, receptionists, administrative assistants, or the "lowly" office staff.
Why?
Their vision isn't clouded by anything other than what is preventing them from getting their job done. They give some of the most honest and forthright answers you're likely to find... Whereas the IT staff likes to play with toys, the execs want to make sure their existing investments are protected, the other managers want to make sure their decisions were right, etc...
They just know that whatever is going in is preventing them from doing there job in a very specific and tangible way.
Everyone's use will be different... For production use, I've rarely found independent reviews that test what I want tested, in the conditions I want, doing the same things I'm looking to do.
For your example case, I'd personally test each product in-house, drawing up conditions and test plans ahead of time. If you're planning a significant deployment, vendors will generally supply product for you to evaluate. Sometimes if you ask nicely, too.
Just my two cents... And yes, I get that it may not be feasible. Its labor and time-intensive. But in-house testing and evaluation almost always beats 3rd party reviews, in my book.
This scenario has always avoided the "pissing match" that inevitably occurs between Dev and IT, in my opinion. Clearly defined roles.
In a previous shop, we treated our in-house developer code releases like any other vendor release. Just because some code was written by the guy down the hall, shouldn't mean that I can't ask for the same level of documentation/support I get from another Tier 1 vendor.
Try downloading Slackware images onto a huge pile of diskettes from a shell server via ZModem over a 14.4 modem and a shitty analog line.
Then having a bad disk, rebooting into DOS, re-dialing the server, and getting that disk again. Not to mention the actual install time.
I would have given my left nut to have that done in under 10 days.
(Having said that, I'm a big fan of Gentoo for single-installation environments. And it took about 3 hours to do my first Gentoo install a few years back, and that was on a Powerbook.)
Despite my dislike of LAN parties, I invested a few bucks in locking screws, and have a case that can't be taken apart without taking said screws out...
Its not the best security, but coupled with a very loud case-intrusion alarm, its not too bad.
This is a start, yeah. You do still have zombie issues, whereby the programs/controllers look up common relay hostnames and/or check Outlook (Express) configurations, but it helps immensely with the spam problem.
Once you do that, the ISP still has problems with users spamming through their relays, but its quite simple for them to detect and auto-block the abusers of that. Or just tiergrub and/or tarpit the high-rate senders, etc...
We've implemented all of this here (I work as a sysadmin at a large ISP), as well as Zero-RDNS blocking, and cut down a LOT of spam.
Hardly. Each entry is hand-verified before its added.
However- if out of, say, 3 million messages recieved from AOL in a certain time period, 30,000 of them from one of my customers IPs come back from AOL as spam (in near real-time), I can be reasonably sure that the message is spam.
This isn't unique to AOL. We do it with a number of other ISPs, as well.
I handle abuse for a very large ISP, and I've found the same thing.
We get a few million reports from AOL per day. Its a lot, but that is the strength of the system.
I process the messages, and based on the number of complaints, I block my customer from sending email though our relay servers (we block port 25/tcp outbound from our largest markets). Stops the spam, and not just to AOL.
This system works very, very well.
I honestly think AOL should be commended for their anti-spam work.
We handle about 14 million incoming messages per day, across 8 qmail-ldap hosts, in a clustered environment. And we use SpamAssassin for mail filtering, as well.
Those 8 hosts (which are quite modest IBM x335 servers) carry almost no load, and their queues are quite small (about 20,000msgs per host, mostly junk waiting to bounce).
The biggest performace increase we saw was when we switched from magnetic disks to Solid State (RAM) disks for the queue drives.
Even if a private security guard were working for someone, and he witnessed (or had a chance to prevent) a crime in progress, he would still be responsible.
Sure, the wording is crappy of the notices, but it seems like this ISP is overreacting.
Take a look here for a sample. This is what the ISP received, most likely.
Since you own this IP address, we request that you immediately do
the following:
....
There really isn't anything that MediaForce can do. Their issue is with the user, not the ISP. I'm sure the ISP will be willing to answer any subpoena requests, but the idea of the ISP seriously thinking about disabling a customer because of one of these notices is silly.
Ask anyone who is in the right place at a big ISP. They'll tell you they get hundreds of these.
I've dealt with JNCS for quite a while, for both work and personal stuff.
Their bundles are a real time saver, prices quite decent, but the best thing, by far, are their return policies. I've rarely seen better.
I had an old Cyrix board (go ahead, laugh) I bought from him a few years ago that died after about a year of being in production. I called them up, and they had me bring it in (they're local to me, but would have taken it by mail, as well). They replaced the board with a PII, as they didn't have the equivalent Cyrix board any more. For no cost. Without even testing it. Just took the return.
Slashdot Mirror
Can't everyone just get back to what they're best at: making truckloads of cash?
I mean, Intel beat AMD in the desktop micro space, and NVIDIA beat ATI in the desktop GPU market. The two losers even had to join forces to stay afloat.
Seems to me that everyone would be happier if they just played nice, and counted their money.
However, I'd think that load balancing those connections would have been the primary issue. Keeping a stateful IMAP connection open across a given cluster of servers can be tricky, versus a straight "wham, bam, thank you ma'am" POP login.
This would generally be a problem only for mere mortals, but Google is, well... Google. So I'm not surprised to see them offering it, even if it is a little late in coming.
I've worked in many different environments, both as a staff member and as an outside consultant.
After the "real" IT staff, who do I go chat with about a given system's issues? The secretaries, receptionists, administrative assistants, or the "lowly" office staff.
Why?
Their vision isn't clouded by anything other than what is preventing them from getting their job done. They give some of the most honest and forthright answers you're likely to find... Whereas the IT staff likes to play with toys, the execs want to make sure their existing investments are protected, the other managers want to make sure their decisions were right, etc...
They just know that whatever is going in is preventing them from doing there job in a very specific and tangible way.
Everyone's use will be different... For production use, I've rarely found independent reviews that test what I want tested, in the conditions I want, doing the same things I'm looking to do.
For your example case, I'd personally test each product in-house, drawing up conditions and test plans ahead of time. If you're planning a significant deployment, vendors will generally supply product for you to evaluate. Sometimes if you ask nicely, too.
Just my two cents... And yes, I get that it may not be feasible. Its labor and time-intensive. But in-house testing and evaluation almost always beats 3rd party reviews, in my book.
This scenario has always avoided the "pissing match" that inevitably occurs between Dev and IT, in my opinion. Clearly defined roles.
In a previous shop, we treated our in-house developer code releases like any other vendor release. Just because some code was written by the guy down the hall, shouldn't mean that I can't ask for the same level of documentation/support I get from another Tier 1 vendor.
Try downloading Slackware images onto a huge pile of diskettes from a shell server via ZModem over a 14.4 modem and a shitty analog line.
Then having a bad disk, rebooting into DOS, re-dialing the server, and getting that disk again. Not to mention the actual install time.
I would have given my left nut to have that done in under 10 days.
(Having said that, I'm a big fan of Gentoo for single-installation environments. And it took about 3 hours to do my first Gentoo install a few years back, and that was on a Powerbook.)
Yep, I bought one at the CompUSA here in Rochester, NY.
The guy looked at me, and said, "You know it doesn't run Windows, right?"
I politely asked him to just ring me up.
There is a free AAC plugin available from the Live marketplace that allow for playback of AAC files.
It works perfectly fine.
http://funnyfox.org.nyud.net:8090/
Hehe... Gran Turismo 2....
God damn that country scares me.
I'm an American and my government scares me...
I'm also not proud of my government, and the fact that people actually voted for the Theocracy we have now in place.
These are some of the darkest days in the history of our country, and we let it happen...
Very interesting times we live in.
Despite my dislike of LAN parties, I invested a few bucks in locking screws, and have a case that can't be taken apart without taking said screws out...
Its not the best security, but coupled with a very loud case-intrusion alarm, its not too bad.
-Adam
I'll believe it when I see it...
Remember, CRM114 was supposed to be the sh*t, too..
This is a start, yeah. You do still have zombie issues, whereby the programs/controllers look up common relay hostnames and/or check Outlook (Express) configurations, but it helps immensely with the spam problem.
Once you do that, the ISP still has problems with users spamming through their relays, but its quite simple for them to detect and auto-block the abusers of that. Or just tiergrub and/or tarpit the high-rate senders, etc...
We've implemented all of this here (I work as a sysadmin at a large ISP), as well as Zero-RDNS blocking, and cut down a LOT of spam.
Hardly. Each entry is hand-verified before its added.
However- if out of, say, 3 million messages recieved from AOL in a certain time period, 30,000 of them from one of my customers IPs come back from AOL as spam (in near real-time), I can be reasonably sure that the message is spam.
This isn't unique to AOL. We do it with a number of other ISPs, as well.
We get a few million reports from AOL per day. Its a lot, but that is the strength of the system.
I process the messages, and based on the number of complaints, I block my customer from sending email though our relay servers (we block port 25/tcp outbound from our largest markets). Stops the spam, and not just to AOL.
This system works very, very well.
I honestly think AOL should be commended for their anti-spam work.
Can be found here.
Need, have...
We had 8 boxes before we went to SSDs. Things were a little chuggy. We replaced the queue disks, and kept the number of machines.
Speed improved, and we can now handle a much larger message load on these boxes, without lifting a finger.
We handle about 14 million incoming messages per day, across 8 qmail-ldap hosts, in a clustered environment. And we use SpamAssassin for mail filtering, as well.
Those 8 hosts (which are quite modest IBM x335 servers) carry almost no load, and their queues are quite small (about 20,000msgs per host, mostly junk waiting to bounce).
The biggest performace increase we saw was when we switched from magnetic disks to Solid State (RAM) disks for the queue drives.
This is good info... Always be sure to read the docs fully before saying X feature doesn't exist in Y product.
I've found the Linux NFS client is rock-solid... Even under really, really heavy load.
My NFS environment runs across a gigabit ethernet backend network, strictly for NFS and inter-server communications.
I mount about 6 terabytes of storage on 70+ machines (yes, each machine sees all that storage), from an EMC Celerra.
The clients are IBM xSeries 1U servers, running RH 7.3, kernel 2.4.18.
The tricks have been mandatory use of NTP, quality network cards and cabling, and redundant Cisco Catalyst switches.
Ada
I'd have to agree with Michael.
Even if a private security guard were working for someone, and he witnessed (or had a chance to prevent) a crime in progress, he would still be responsible.
I think. But thats just my moral compass talking.
Here's a nice link with more background on it.
http://www.chillingeffects.org/piracy/notice.cgi?N oticeID=386
Take a look here for a sample. This is what the ISP received, most likely.
Since you own this IP address, we request that you immediately do the following:
There really isn't anything that MediaForce can do. Their issue is with the user, not the ISP. I'm sure the ISP will be willing to answer any subpoena requests, but the idea of the ISP seriously thinking about disabling a customer because of one of these notices is silly.
Ask anyone who is in the right place at a big ISP. They'll tell you they get hundreds of these.
Adam
I've dealt with JNCS for quite a while, for both work and personal stuff.
Their bundles are a real time saver, prices quite decent, but the best thing, by far, are their return policies. I've rarely seen better.
I had an old Cyrix board (go ahead, laugh) I bought from him a few years ago that died after about a year of being in production. I called them up, and they had me bring it in (they're local to me, but would have taken it by mail, as well). They replaced the board with a PII, as they didn't have the equivalent Cyrix board any more. For no cost. Without even testing it. Just took the return.
Now thats service.