How is that a problem with GMail? Seems to me it's a problem with sorbs.
It is a problem with neither. The recipient (3rd party) that uses SORBS bounces the mail from GMail because their criteria are way too blunt (one blackhole and blocked). In practice however this becomes a problem for the GMail user. I see this happening too often for my liking. For more info check with the Google Group.
Below is a reaction from SORBS on the subject which I took from the newsgroups:
Unfortunately Gmail is taking no action to prevent the spam and instead expects the rest of the world to play spammer whack-a-mole each time their spamming users sign up for a new address and send the identical spam. Gmail needs to take action to prevent the spam from leaving their network, not expect the rest of the world to police their users for them. As long as spam is continuing to come from Gmail IPs, they will be listed by SORBS.
Dan
SORBS Volunteer
Apparently GMail and SORBS have a disagreement on the subject...
One of the main problems with GMail is the "on behalf of" thing when trying to masquerade under a valid alternative email address.
I'd say one of the mail problems with GMail is the fact that their outbound SMTP relayers are off-and-on listed in the dnsbl.sorbs.net blackhole. This means mail you send out may get blocked by receiving servers that check this blackhole.
I'm regularly getting these kinds of messages when I send out mail and that really sucks:
Perhaps my example is somewhat broken, it however illustrates that a decent JVM can keep up with (or even outperform) a (perhaps not all) similar function of a C compiled product. And since in my example it applies to most of the functions of the Java product in comparison to the C product (i.e. serving HTTP and the like) I dare to state: "Don't underestimate the execution-speed of Java in a decent JVM.". Nothing more, nothing less.
I think the OS is the very, very last thing that Google would aim for. They'd go for all the application and framework space first.
I doupt they'll ever go there (OS). The browser is their OS, and "The Network is the Computer(tm)". They are basically battling Microsoft with the monopolists own precious browser. There is nothing Microsoft van do to stop Google from 'running' on its OS. Is that irony or what?
...stops spam! I've been using it for some time now (server-side) and it has been so acurate that I have proceded to move alleged spam to/dev/null.
Mail that Spamassassin thinks is spam but isn't can't be that interesting anyway.
Re:Cool! Just like form AutoComplete
on
Google Suggest
·
· Score: 1
> it's bound to be pretty computationally intensive.
Only searching previously entered search strings might not be that bad. 80-20 rule probably applies here too.
I've been using Resin for a couple of years now. Without the caching and clustering since I have my own caching mechanism and don't need clustering.
That said I can recommend Resin to anyone (that needs an app-server or a servlet container) for its performance and (now with the new Linux NPTL threads) also scalability. I've been making websites with Java (first on OS/2) since 1998 and have used quite a few app servers until I ran into Resin 2 or 3 years ago. Haven't thought about looking any further since.
BTW, since I'm a paying customer I have no idea what they mean by "performance capabilities"
I reckon that there are also stability and other issues besides the much heard about Internet vulnerabilities that need patching. Like fileserver crashing, or the database corrupting data. Understandable that all the different (specialized) software must be tested against the OS patches.
This (regression) testing would also be the case on other OS'es I assume.
Eventhough the source benefits highly from its own conclusion they do have a point. I don't think it's wise to download the latest (e.g.) Fedora ISOs and run a nuclear launch site with it. On the other hand, it should be entirely possible to certify Linux secure in one way or the other, thus having a 'secure' internal/military branch of the sources that no-one from outside can contribute to.
Anyway, even GreenHills could program a backdoor in their software...
I live in the EU, and the EU doesn't seem to agree on anything. I wish it were true that the EU is a superpower in 50 years, but we'll probably still be drafting the EU-constitution and letting the voters kill the draft with a referendum.
China on the other hand will be the greatest economy long before 50 years have gone by. Invest in the Chinese stock-market!
Your story works 2 ways. Hooray for you if you think Windows covers your needs. Other people feel the same way about Linux (fashionable or not) or OSX et.al. Just because Windows covers your needs doesn't mean that is true for everyone (and vice-versa).
They tried this with OS/2 when Windows95 came out. Unfortunately with all the resources they spent, almost nobody switched, and believe me that OS/2 2.1 was a lot better dan Win95. I did use it (server and client), until I switched to Linux 3 or 4 years ago.
What I don't understand is that Tannenbaum is proud of Sun and Apache for processing 150K hits a day. I run a system where a single Linux PC processes 6 to 10 Mio hits a day. This being calls to a servlet that interacts with a RDBMS for each hit.
Not wanting to be offtopic, this illustrates for me the academic view on the world vs. the real view on the world. AFAIK a moderately powerful SUN server with Apache should be able te handle 150K hits a minute when requesting a static page.
Slashdot Mirror
Seriously, this is what we'll have in our desktop computer for $150 in at most 5 years.
Geesh, now you're exaggerating! It's not that big a cock.
But it sure is pink!
It is a problem with neither. The recipient (3rd party) that uses SORBS bounces the mail from GMail because their criteria are way too blunt (one blackhole and blocked). In practice however this becomes a problem for the GMail user. I see this happening too often for my liking. For more info check with the Google Group.
Below is a reaction from SORBS on the subject which I took from the newsgroups:
Apparently GMail and SORBS have a disagreement on the subject...
I'd say one of the mail problems with GMail is the fact that their outbound SMTP relayers are off-and-on listed in the dnsbl.sorbs.net blackhole. This means mail you send out may get blocked by receiving servers that check this blackhole.
I'm regularly getting these kinds of messages when I send out mail and that really sucks:
PERM_FAILURE: SMTP Error (state 9): 554 Service unavailable; Client host [64.233.166.180] blocked using dnsbl.sorbs.net; Spam Received See: http://www.sorbs.net/lookup.shtml?64.233.166.180
Perhaps my example is somewhat broken, it however illustrates that a decent JVM can keep up with (or even outperform) a (perhaps not all) similar function of a C compiled product. And since in my example it applies to most of the functions of the Java product in comparison to the C product (i.e. serving HTTP and the like) I dare to state: "Don't underestimate the execution-speed of Java in a decent JVM.". Nothing more, nothing less.
Don't underestimate the execution-speed of Java in a decent JVM. For example: My Java-based HTTPD outruns Apache HTTPD for static file serving.
How about 4DOS ?
...stops spam! I've been using it for some time now (server-side) and it has been so acurate that I have proceded to move alleged spam to /dev/null.
Mail that Spamassassin thinks is spam but isn't can't be that interesting anyway.
> it's bound to be pretty computationally intensive. Only searching previously entered search strings might not be that bad. 80-20 rule probably applies here too.
Gee, you must really be a blast at parties with such a deep sense of humor. Ever heard of terms like sarcasm and irony?
Well, since it is open-source now you may want to remove those random Thread.sleep() calls yourself.
I've been using Resin for a couple of years now. Without the caching and clustering since I have my own caching mechanism and don't need clustering.
That said I can recommend Resin to anyone (that needs an app-server or a servlet container) for its performance and (now with the new Linux NPTL threads) also scalability. I've been making websites with Java (first on OS/2) since 1998 and have used quite a few app servers until I ran into Resin 2 or 3 years ago. Haven't thought about looking any further since.
BTW, since I'm a paying customer I have no idea what they mean by "performance capabilities"
I reckon that there are also stability and other issues besides the much heard about Internet vulnerabilities that need patching. Like fileserver crashing, or the database corrupting data. Understandable that all the different (specialized) software must be tested against the OS patches. This (regression) testing would also be the case on other OS'es I assume.
Eventhough the source benefits highly from its own conclusion they do have a point. I don't think it's wise to download the latest (e.g.) Fedora ISOs and run a nuclear launch site with it. On the other hand, it should be entirely possible to certify Linux secure in one way or the other, thus having a 'secure' internal/military branch of the sources that no-one from outside can contribute to.
Anyway, even GreenHills could program a backdoor in their software...
I live in the EU, and the EU doesn't seem to agree on anything. I wish it were true that the EU is a superpower in 50 years, but we'll probably still be drafting the EU-constitution and letting the voters kill the draft with a referendum.
China on the other hand will be the greatest economy long before 50 years have gone by. Invest in the Chinese stock-market!
Your story works 2 ways. Hooray for you if you think Windows covers your needs. Other people feel the same way about Linux (fashionable or not) or OSX et.al. Just because Windows covers your needs doesn't mean that is true for everyone (and vice-versa).
My parent-message is modded as "Funny" but should have been modded "Sad" as this is indeed the bitter irony.
They tried this with OS/2 when Windows95 came out. Unfortunately with all the resources they spent, almost nobody switched, and believe me that OS/2 2.1 was a lot better dan Win95. I did use it (server and client), until I switched to Linux 3 or 4 years ago.
What I don't understand is that Tannenbaum is proud of Sun and Apache for processing 150K hits a day. I run a system where a single Linux PC processes 6 to 10 Mio hits a day. This being calls to a servlet that interacts with a RDBMS for each hit.
Not wanting to be offtopic, this illustrates for me the academic view on the world vs. the real view on the world. AFAIK a moderately powerful SUN server with Apache should be able te handle 150K hits a minute when requesting a static page.