Michael- will do, but one question-will these changes have any effect on my other online activities? I rarely do any significant uploads, but download rather promiscuously.
Are you calling your computer a slut? LOL
These change in these settings will conceivably have some effect upon u/l or d/l, couldn't say what that might be. Internet details are off or at the edge of my horizon of knowledge. I'd suggest going to a speed-measuring website, PCPitstop comes to mind, though it requires you to use Internet Explorer to access the test phase, because the site makes use of ActiveX controls.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
My input is based on what I've read, and what I've tried on the Mac side... but the default of 1500 is usually the problem. 1400 solves it for many people. 576 is the "other type packet" size and supposedly would solve "this particular problem" for anyone. Also realize that there are many _places_ to change packet sizes; packets can get, uh, repackaged, at routers, cable/dsl modems, wireless cards, inside your ISP... so setting the value at the computer _should_ mean it won't get _bigger_ as it travels, but there seem to be no guarantees.
Going below 576... I don't think I would. If that doesn't fix it, I'd go back to 1500 and look for another answer. But realize that my network knowledge is at the "this wire goes where?" level.
As for impact on other things, yes, it can slow things down a bit (probably not much, but more and more as you go smaller) because you'll be sending more packets to transfer the same amount of information. Thus this is ideally a "temporary" fix, only until we find which internet carrier has messed things up, and get them to fix it. (Or more likely they stumble on the problem themselves.)
@Michael: Is there _really_ no easier way, in some control panel or something, to change this on Windows? I'll admit it was hard to find, buried many "advanced" tabs and clicks deep, but I finally found a field where I could just type in a new value on the Mac.
@Michael: Is there _really_ no easier way, in some control panel or something, to change this on Windows? I'll admit it was hard to find, buried many "advanced" tabs and clicks deep, but I finally found a field where I could just type in a new value on the Mac.
Bill,
I don't know of a way within Windows, but that hardly means that such a way does not exist. As I told Brad, this stuff is at or past the limits of my personal horizon. I didn't even know where to find the registry fix, just that those TCP/IP settings had to be in there somewhere, so I searched it for "mtu", skipped through all the false leads where that string appears in other stuff, and finally found the setting within that folder. System Mechanic, the utility I suggested to Brad, makes adjusting the settings super-easy, a simple UI. I'm much more comfortable advising someone to use it than to have them go rooting through the registry, which is like break-dancing through a minefield.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
@Michael: Is there _really_ no easier way, in some control panel or something, to change this on Windows? I'll admit it was hard to find, buried many "advanced" tabs and clicks deep, but I finally found a field where I could just type in a new value on the Mac.
Bill,
I don't know of a way within Windows, but that hardly means that such a way does not exist. As I told Brad, this stuff is at or past the limits of my personal horizon. I didn't even know where to find the registry fix, just that those TCP/IP settings had to be in there somewhere, so I searched it for "mtu", skipped through all the false leads where that string appears in other stuff, and finally found the setting within that folder. System Mechanic, the utility I suggested to Brad, makes adjusting the settings super-easy, a simple UI. I'm much more comfortable advising someone to use it than to have them go rooting through the registry, which is like break-dancing through a minefield.
Hi guys-taking lunch, and have been receiving same error messages as before-many many times. By the way, I have 3 control sets in that hive, the MTU value appears in 2 and 3, and System Mechanic only changed #2. Think i will just try running the prog again from time to time and watch for news of an answer.
Makes me hesitate to add another project. Is this kind of baby-sitting typical?
Hi guys-taking lunch, and have been receiving same error messages as before-many many times. By the way, I have 3 control sets in that hive, the MTU value appears in 2 and 3, and System Mechanic only changed #2. Think i will just try running the prog again from time to time and watch for news of an answer.
Makes me hesitate to add another project. Is this kind of baby-sitting typical?
(I did manually change both control sets to 426, but for others relied on System Mechanic. If my patience returns will re-try some other values.
Makes me hesitate to add another project. Is this kind of baby-sitting typical?
Brad,
I'd be hesitant to add another project right away, too. As far as I can tell, the original theory of the troubleshooters was that it was something to do with the Solaris servers, because Einstein, which alone (I think) uses Apache servers, seemed to be immune to the problem. The conclusion that they seem to be coming to now is that it is a result of a widespread internet problem, and they are trying to identify which IP carrier(s) is the "culprit".
microcraft
"The arc of history is long, but it bends toward justice" - MLK
Michael- Don't know if theres any point, but I will contact my ISP to see if they have a clue. And will eventually get around to trying some other MTU values. Exasperating cuz I logged almost 7000 hrs on old seti with no trouble. Notice my join date- there was trouble then also, don't recall what.
I've been doing a little bit of research, and come up with some material. First, how deep do you want to go into this? You seem somewhat "fearless" about delving into the depths, or I wouldn't bother you with this.
Here is an article on annoyances.org about MTU settings.
And here is a piece about a DOS prompt program named "tracert", which can be useful in diagnosing connection problems. It also has a link to a page about "pathping", another diagnostic tool. If you decide to run these, you can post the results here, and see if our resident net guru can get a look at them, though I have to tell you in advance that I'm sure he is and has been up to his neck in trying to resolve things at the server end, so who knows when he'll be able to tackle it here.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
RE: Michael- will do, but
)
Are you calling your computer a slut? LOL
These change in these settings will conceivably have some effect upon u/l or d/l, couldn't say what that might be. Internet details are off or at the edge of my horizon of knowledge. I'd suggest going to a speed-measuring website, PCPitstop comes to mind, though it requires you to use Internet Explorer to access the test phase, because the site makes use of ActiveX controls.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
Michael- ok-456 still fails,
)
Michael- ok-456 still fails, but returns different report:
12/6/2005 10:07:40 AM||Starting BOINC client version 5.2.13 for windows_intelx86
12/6/2005 10:07:40 AM||libcurl/7.14.0 OpenSSL/0.9.8 zlib/1.2.3
12/6/2005 10:07:40 AM||Data directory: C:\\Program Files\\BOINC
12/6/2005 10:07:40 AM||Processor: 1 AuthenticAMD AMD Athlon(tm) XP 2200+
12/6/2005 10:07:40 AM||Memory: 479.48 MB physical, 1.42 GB virtual
12/6/2005 10:07:40 AM||Disk: 69.24 GB total, 36.77 GB free
12/6/2005 10:07:40 AM|Einstein@Home|Computer ID: 465356; location: home; project prefs: default
12/6/2005 10:07:40 AM||General prefs: from Einstein@Home (last modified 2005-05-30 09:38:31)
12/6/2005 10:07:40 AM||General prefs: no separate prefs for home; using your defaults
12/6/2005 10:07:41 AM||Remote control not allowed; using loopback address
12/6/2005 10:07:41 AM|Einstein@Home|Started upload of l1_0110.0__0110.0_0.1_T01_S4lD_3_0
12/6/2005 10:07:44 AM|Einstein@Home|Temporarily failed upload of l1_0110.0__0110.0_0.1_T01_S4lD_3_0: error 500
12/6/2005 10:07:44 AM|Einstein@Home|Backing off 1 hours, 3 minutes, and 38 seconds on upload of file l1_0110.0__0110.0_0.1_T01_S4lD_3_0
So I guess its wait and see. Leaving for work soon, so will let it stew. And speed tests seem within norm, but my cable is wildly inconsistent.
Brad
"You have confused the true with the real."
My input is based on what
)
My input is based on what I've read, and what I've tried on the Mac side... but the default of 1500 is usually the problem. 1400 solves it for many people. 576 is the "other type packet" size and supposedly would solve "this particular problem" for anyone. Also realize that there are many _places_ to change packet sizes; packets can get, uh, repackaged, at routers, cable/dsl modems, wireless cards, inside your ISP... so setting the value at the computer _should_ mean it won't get _bigger_ as it travels, but there seem to be no guarantees.
Going below 576... I don't think I would. If that doesn't fix it, I'd go back to 1500 and look for another answer. But realize that my network knowledge is at the "this wire goes where?" level.
As for impact on other things, yes, it can slow things down a bit (probably not much, but more and more as you go smaller) because you'll be sending more packets to transfer the same amount of information. Thus this is ideally a "temporary" fix, only until we find which internet carrier has messed things up, and get them to fix it. (Or more likely they stumble on the problem themselves.)
@Michael: Is there _really_ no easier way, in some control panel or something, to change this on Windows? I'll admit it was hard to find, buried many "advanced" tabs and clicks deep, but I finally found a field where I could just type in a new value on the Mac.
RE: @Michael: Is there
)
Bill,
I don't know of a way within Windows, but that hardly means that such a way does not exist. As I told Brad, this stuff is at or past the limits of my personal horizon. I didn't even know where to find the registry fix, just that those TCP/IP settings had to be in there somewhere, so I searched it for "mtu", skipped through all the false leads where that string appears in other stuff, and finally found the setting within that folder. System Mechanic, the utility I suggested to Brad, makes adjusting the settings super-easy, a simple UI. I'm much more comfortable advising someone to use it than to have them go rooting through the registry, which is like break-dancing through a minefield.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
RE: RE: @Michael: Is
)
Hi guys-taking lunch, and have been receiving same error messages as before-many many times. By the way, I have 3 control sets in that hive, the MTU value appears in 2 and 3, and System Mechanic only changed #2. Think i will just try running the prog again from time to time and watch for news of an answer.
Makes me hesitate to add another project. Is this kind of baby-sitting typical?
"You have confused the true with the real."
Hi guys-taking lunch, and
)
Hi guys-taking lunch, and have been receiving same error messages as before-many many times. By the way, I have 3 control sets in that hive, the MTU value appears in 2 and 3, and System Mechanic only changed #2. Think i will just try running the prog again from time to time and watch for news of an answer.
Makes me hesitate to add another project. Is this kind of baby-sitting typical?
(I did manually change both control sets to 426, but for others relied on System Mechanic. If my patience returns will re-try some other values.
"You have confused the true with the real."
RE: Makes me hesitate to
)
Brad,
I'd be hesitant to add another project right away, too. As far as I can tell, the original theory of the troubleshooters was that it was something to do with the Solaris servers, because Einstein, which alone (I think) uses Apache servers, seemed to be immune to the problem. The conclusion that they seem to be coming to now is that it is a result of a widespread internet problem, and they are trying to identify which IP carrier(s) is the "culprit".
microcraft
"The arc of history is long, but it bends toward justice" - MLK
Michael- Don't know if theres
)
Michael- Don't know if theres any point, but I will contact my ISP to see if they have a clue. And will eventually get around to trying some other MTU values. Exasperating cuz I logged almost 7000 hrs on old seti with no trouble. Notice my join date- there was trouble then also, don't recall what.
"You have confused the true with the real."
Brad, I've been doing a
)
Brad,
I've been doing a little bit of research, and come up with some material. First, how deep do you want to go into this? You seem somewhat "fearless" about delving into the depths, or I wouldn't bother you with this.
Here is an article on annoyances.org about MTU settings.
And here is a piece about a DOS prompt program named "tracert", which can be useful in diagnosing connection problems. It also has a link to a page about "pathping", another diagnostic tool. If you decide to run these, you can post the results here, and see if our resident net guru can get a look at them, though I have to tell you in advance that I'm sure he is and has been up to his neck in trying to resolve things at the server end, so who knows when he'll be able to tackle it here.
microcraft
"The arc of history is long, but it bends toward justice" - MLK
Have the same difficulties
)
Have the same difficulties under XPhome connecting via DSL