how-to block ads
|
dellsweig
Extreme Aerobatics
Premium,MVM
join:2003-12-10
Campbell Hall, NY
 ·Time Warner VOIP
·Vonage
 ·RoadRunner Cable
2 edits | reply to dellsweig
Re: Looking for the Common Thread in router problems
Just to goive you guys an update - not want you want to hear either (knock on wood)
The router has been fine since the beginning of December. As you know, both this and my old V1 started freezing once a day - sometimes more often. I replaced the V1 with the V4, upgraded software and started playing with settings.
The device runs 24/7/365. Speed on Road runner 2800/360 average.
UPnP off
VPN passsthrough enabled
Logging off
WEP/WAP off
MTU 1400
no cloned MAC
2 forwarded ports (5060-5062) to valid STATIC addresses
Port filtering on 3 ranges
WAN access disabled
remote upgrade enabled
DHCP client lease 32767 (6 dhcp client PC,s 2 hardwire static IP cisco ATA)
Filter multicast - disabled
SSID enabled - channel 6
Obtain IP automatically
BTW: My ISP does NOT block any of the known service ports (1000 or less) and most likely does not block any ports
I am not offereing solutions here - just observations. Both Linksys and my ISP said to change the MTU. I did and the problem cleared. I set it back (default) and the problem occurs again. I am sure the combination of other settings help in some way (buffer usage, etc). This experiment is repeatable - tech support at my ISP asked me to repeat it.
The major chains and ISP's are distributing these boxes. I doubt a profit making organization like Best Buy would continue to sell a problem SKU that flooded their return or service desk. | |
ChrisDAT
Google Keyword Compsysnyc
join:2002-02-26
Hollis, NY
|  reply to dellsweig
biker -- You hit the nail on the head about the debug and stuff, and thanks for the info. I also learned long ago, that before I post anything of substance to "copy" to the clipboard, FIRST -- I feel your pain.
It's important to note that worm attacks are random -- in one hour you could see 100 hits, and for the next day, you may not see any. Some of the attacks are also "smart" enough that when they find a good target IP, they remember and propagate that info about the "soft target," so that an IP change can totally change the nature of the issue -- If you have a static IP, you are probably more succeptible -- keep in mind, these little devils are designed by people who know this stuff, they certainly know that without a dynamic IP [this is usually an ISP-wide setting], you're a "sitting duck" in a way. I also believe that of the "well known" exploits, there are probably twice as many "unknown" exploits because, once again, the bad guys are always one step ahead of the good guys, they also follow the CERTs and the workarounds, and immediately go to work on an attack for the workaround, most likely before you even see it. It may only take ONE bad [malformed] packet (with an invalid MTU or MSS field possibly) to jack the router. I would be interested in hearing from RonS on this issue.
More people need to report in this....
dells, You are also correct -- We cannot get to the bottom of this issue if our basic settings are "off" from the start -- LinkSys boxes do not do Max Path MTU Discovery [a feature that may solve many issues], while PCs have this turned ON by default, moreover the router's MTU must be less than or equal to the MTU of your ISP [actually, I think it is the modems that are goofy on that one]. The problem still persists for some people regardless of the setting of the MTU -- moreover, the lockups usually become less frequent because the MTU issue is solved when you "get it right." The MTU and MSS are related in that they determine the "fields" of the packet record which in turn tells firewalls and applications how to interpret the packet data. Routers along the way do not examine this info, so it is transparent to them -- that's what makes malformed packets "explosive" only at the end point -- the smart bombs of the worm world, they are usually directed at one specific application.
I really wish I could "look inside" the LinkSys better, in the absence of realtime information about the operating environment of the router, we can only look at the external conditions and make a best guess based on those observations as to how the router is responding to/effected by them. If we're lucky and look deep enough a "pattern" will emerge. | |
dellsweig
Extreme Aerobatics
Premium,MVM
join:2003-12-10
Campbell Hall, NY
·Time Warner VOIP
·Vonage
·RoadRunner Cable
| Thanks Chris - good post!!!
I will say how suprised I was that setting my MTU back to default - re-introduced the problem!! That was the first time the router had been re-set in weeks. A very helpful tech at my ISP wanted to try it - actually what a concept - a cable ISP tech that would talk to me about a router .
Thinking about it - the fact that this problem started after some specific date/event maybe suggests a firmware upgrade in some part of the ISP network.... Now if we could only get that kind of info....
Hey - you have to try those 5.5 Db antennas from Radioshack - I see about a 20% difference on the Signal graphs and get coverage deep into the corners of my house now!! | |
big greg
Premium,MVM,Ex-Mod 2005-6
join:2003-10-11
Boston, MA
clubs: 
| reply to biker45
said by biker45  :
As a software developer, I know how I'd solve this problem. I'd write a debug version of the firmware that traced every significant event and write the trace data to the log. Then, I'd ask for volunteers (with plenty of space on their hard drives) to run the debug version til their router hangs, and send me the log. It would only take a few logs, cold pizza, and some coffee to work through the logs and find the bug (er I mean cause of the hang).
I'm a software developer too, and clearly that's the way to attack this problem.
said by biker45 :
I just do not think Linksys is going to solve this problem. They are going to put their resources into the 802.11g product line, and the old BEFW11S4 will die a slow (and unsupported) death. I am thankful (and very surprised) that they actually provided firmware for the BEFW11S4 that supports WPA (but when I enabled WPA on my BEFW11S4, I found out that my wireless adapter (WUSB11 v2.8) does not support WPA, even though Linksys says it does).
I can give you one reason that the 1.50 code supports WPA: marketing. They can sell more of these crippled boxes if they claim they support WPA.
Remember we all got a bottom of the line device here. People go to the mall and buy the cheapest wireless router they can find. The V1/2/3 boxes were cheap (but reliable) but the v4 box is most likely far cheaper to make. Cheaper to make results in more profit.
Linksys could care less that this horrible product is tarnishing their reputation... I'm sure sales of this POS are not hurt one bit by it's flaws. People just buy the cheapest one (I did) or they purchased one based on experience with earlier models (Mom's V3 is rock solid) or because they saw the Cisco logo on the box (marketing ploy).
If I was in the department where the 1.50 code was produced, I'd be pretty ashamed of myself. Clearly the code has had major changes in the 1.50 release. Management rushed the code out because they wanted it to claim that it supports WPA. Now that it's out, the engineers have no doubt moved to other products.
The priority at Linksys is to move hardware at the expense of their reputation. | |
RonS
Madm0nke
Premium
join:2000-06-19
Dayton, OH
| reply to ChrisDAT
said by ChrisDAT :
It's important to note that worm attacks are random -- in one hour you could see 100 hits, and for the next day, you may not see any. Some of the attacks are also "smart" enough that when they find a good target IP, they remember and propagate that info about the "soft target," so that an IP change can totally change the nature of the issue -- If you have a static IP, you are probably more succeptible -- keep in mind, these little devils are designed by people who know this stuff, they certainly know that without a dynamic IP [this is usually an ISP-wide setting], you're a "sitting duck" in a way. I also believe that of the "well known" exploits, there are probably twice as many "unknown" exploits because, once again, the bad guys are always one step ahead of the good guys, they also follow the CERTs and the workarounds, and immediately go to work on an attack for the workaround, most likely before you even see it. It may only take ONE bad [malformed] packet (with an invalid MTU or MSS field possibly) to jack the router. I would be interested in hearing from RonS on this issue.
Very interesting and thought inducing post. I read back and some people say that "All the sudden the router just stopped working. It was fine for weeks" People with cable modems keep the same IP address to the MAC address of the device the modem is hooked up to for months. When I put the new router in I got a different IP address. Could it be that the IP address hasn't been used and the worms are ignoring it for now. Maybe eventually they will see the IP in use and send a bad packet to it and start problems for me?
That wouldn't explain the problems for DSL users who usually get a different IP address each time the have to power cycle their Router. It would be interesting if the people having the problem that are on cable modems would unplug the Network cable from the router and plug the modem into their computer. This would get you a different IP address. Then wait a day or two and plug it back into your router. This should get you a different IP address than you had previously (write down your IP address to verify) Then see if the problem is still there.
--
3 out of 4 people make up 75% of the population. | |
ChrisDAT
Google Keyword Compsysnyc
join:2002-02-26
Hollis, NY
|  reply to dellsweig
I have an idea biker, that If you're game, we could try.
Let's assume our "Gremlin" is a malformed port 80 [or some other port or service] packet that chokes the LS when it is recieved...
Your mission, should you choose to accept it, is to tell the router to forward port 80 requests to never-neverland, or better still, a PC, possibly with a running HTTPd, that can properly [and safely] handle this request.
What this would do is force the router to modify the packet, maybe even "fix" it, such that it encounters the LinkSys in a different context. Having an HTTPd to recieve the request would "expose" the actual HTTP request that is being made by viewing the log of the HTTPd. It could be somewhat "risky" to use the MS IIS, because I know many of the exploits are specifically targeted at the MS web server.
Would anyone like to pursue this issue?
Hi Biggie-Gee! You are on the money describing the hardware of the LS -- What we're trying to do here is hot-rod the box a bit so that it can at least meet our expectations that come from knowledge of hardware that costs hundreds of times more than these little boxes. Maybe we will not succeed, but let me put it in context: I have a 1970 Chevrolet Chevelle [red, 396] that has no digital electronics to speak of [ie. no brain] -- when the weather changes more than a little, it needs to be re-tuned or it will not run, if I put cheap gas in it, it will let me know, the paint is so old, it is cracking and appears to fall off as I look at it, and it drinks gasoline like the Exxon Valdiz -- on the other hand, it's an absolute thrill to drive, it gets "thumbs ups" where ever I go, and I can't step out of it without someone asking me "how much." I bought the car for a cool $1000 and have probably put more than 5x that into it since -- will I get rid of it - NEVER. Ultimately, what we want to do is find the limitations of the LinkSys package, so we can work around them, and live with them.
It is absolutely easy to jump ship on LinkSys, as there is a world of similarly-priced and featured boxes out there, but if the problem is what we're working here, it will do no good -- the LinkSys has the features I want and need, and my LinkSys works, if I'm correct here, the problem is not even the fault of LinkSys, and finding this elusive "silver bullet" will apply across brands. | |
big greg
Premium,MVM,Ex-Mod 2005-6
join:2003-10-11
Boston, MA
clubs:
1 edit | reply to RonS
I run a small ISP and when there are attacks, most often they come all over my IP block from IP addresses close to mine. Example: if my IP block was 4.4.4.0/32 then I would often get attacks on random addresses in my subnet from 4.4.1.0/32 through 4.4.8.0/32. This makes a lot of sense, cause if you want to find more live hosts to infect, you don't want to pick totally random IPs. You want to pick some that you know are assigned, connected, and friendly to your advances.
There's one subnet close to mine that must be run by idiots, as they get infected and stay that way for months.
Anyway, if you are on a cable system it would seem most natural to attack your neighbors, some of which you would share a network segment with, as not only are they connected but you can get to them quickly, and perhaps without any telltale traffic going to cable system's internet gateway. Same thing with DSL, as everyone connecting to a particular CO most likely gets an address out of the same subnet.
For those types of worms that like to attack neighbors, changing your IP address would NOT matter much.
I also use a static IP from a T1 for my W41 and it hangs just as reliably as well as you guys on DSL or cable.
edit: added NOT | |
ChrisDAT
Google Keyword Compsysnyc
join:2002-02-26
Hollis, NY
| reply to dellsweig
You see big greg, this is called thinking "outside the box" -- The fact that you get attacks from a T1 takes both the DSL and cable modems out of the picture. Your post also speaks volumes to the fact that the little critters are "out there" and that they MAY have an adverse effect on your LinkSys in particular.
I know you meant to say "For those types of worms that attack neighbors, changing your IP address would [NOT] matter much." However, I would dare to disagree -- while a change in IP would not change the frequency of getting a random attack, it would influence a directed attack.
To the "subnet run by idiots" -- that's possibly a group of PCs that are running unprotected/filtered by a firewall. Most of us spend far less time on outbound filtering than we do on inbound forwarding. Outbound filtering can prevent us from becoming part of the problem.
On my DSL, I usually get the same address after a router or modem reset via a settings change. After an ISP hiccup, I often get a vastly different IP (network addresss wise), even the PPPoE server can be different [my first hop in an outbound tracert], so I don't know if I can "make" my ISP give me a new address without going "dark" [turning off the DSL modem] for a certain time period. From the LogViewer, 99.9% of the stray requests on unblocked ports (overwhelmingly 137,139 and 445, but there are others) come from remote ISPs.
You are dead right about cable, any communications by near neighbors will at least be seen by the cable modem, and possibly passed on to to the router. Cable has that vulnerability, especially if the modem forwards broadcasts and multicasts.
I am concerned that sometimes as we suggest/post possible solutions, a person reads them, applies the knowledge, their problem goes away, and they never revisit the site to say "hey, it worked for me!" The feedback is important, even if it is to say "that s**t didn't even work." Please, everyone, respond. | |
big greg
Premium,MVM,Ex-Mod 2005-6
join:2003-10-11
Boston, MA
clubs:
| said by ChrisDAT :
You see big greg, this is called thinking "outside the box" -- The fact that you get attacks from a T1 takes both the DSL and cable modems out of the picture. Your post also speaks volumes to the fact that the little critters are "out there" and that they MAY have an adverse effect on your LinkSys in particular.
So at least we know it's not settings related to your incoming connection (static, dynamic, pppoe, DSL, cable, T1) so that "rules out" a certain (small) section of the router code, and it "rules in" certain environmental factors encountered on any IP address.
said by ChrisDAT :
I know you meant to say "For those types of worms that attack neighbors, changing your IP address would [NOT] matter much."
Yeah, I should have re-read that.
said by ChrisDAT :
However, I would dare to disagree -- while a change in IP would not change the frequency of getting a random attack, it would influence a directed attack.
True. I was referring to the random critter attack, the kind that sprays all my IPs with the same junk. In that case changing your IP address to another in the same subnet would NOT matter that much.
Certainly a directed attack on a certain live IP is a whole other ball game.
said by ChrisDAT :
To the "subnet run by idiots" -- that's possibly a group of PCs that are running unprotected/filtered by a firewall. Most of us spend far less time on outbound filtering than we do on inbound forwarding. Outbound filtering can prevent us from becoming part of the problem.
Yeah over there at typhoid mary corporation there a number of unpatched Windows 98 machines, and a bunch of other stuff. At least they shut down the unpatched NT 4.0 machine. 
said by ChrisDAT :
On my DSL, I usually get the same address after a router or modem reset via a settings change. After an ISP hiccup, I often get a vastly different IP (network addresss wise), even the PPPoE server can be different [my first hop in an outbound tracert], so I don't know if I can "make" my ISP give me a new address without going "dark" [turning off the DSL modem] for a certain time period.
The DHCP server has remembered your MAC address in its DHCP client table. When the modem comes online it asks for another address and the DHCP server says aha, renewal for this MAC address.
When the ISP bounces the DHCP server, it doesn't have your MAC address any more. So the cable modem asks for an address and in all likelyhood gets a new one.
said by ChrisDAT :
You are dead right about cable, any communications by near neighbors will at least be seen by the cable modem, and possibly passed on to to the router. Cable has that vulnerability, especially if the modem forwards broadcasts and multicasts.
Want I meant was if there is traffic between two IPs that are in the same routing subnet, those packets will NOT be passed to core of the cable network. The source and destination is the local subnet. Not only that you have high speed access for brute attacks on other IPs in your subnet. 
So the cable network would have no way of seeing this traffic. They can kick the customer off the cable system, but they won't know he's attacking your system.
said by ChrisDAT :
I am concerned that sometimes as we suggest/post possible solutions, a person reads them, applies the knowledge, their problem goes away, and they never revisit the site to say "hey, it worked for me!" The feedback is important, even if it is to say "that s**t didn't even work." Please, everyone, respond.
OK, I've got an idea that maybe hasn't been tried. I'm going to put the router behind another NAT server rather than on the unwashed Internet. (I wanted to keep it on the Internet segment in case my WEP was broken... they can do whatever they want out there, but keep them the hell off the internal network at any cost.)
Yes, little Linky will get a nice quiet relaxing cozy private little network segment. All protected from the outside world. That will tell us if it is the network traffic, or not. I can set it up tonight. I'll let you know the results. | |
biker45
join:2003-10-18
Erie, CO
| reply to dellsweig
ChrisDAT: Good idea about forwarding port 80 requests to never-land. Let's wait to hear from big greg's experiment in isolating his LS from the outside world. Thanks big greg for the effort!!
I think ChrisDAT's observation is right on. Based on the posts in this thread, most of us here are more interested in finding a solution to this problem because of the challenge, rather than trading in our Linksys hardware for something that works. One thing's for certain, when we solve this one, Linksys has many more problems (er I mean challenges) to work on 
| |
qworster
join:2001-11-25
Los Angeles, CA
 ·DSL EXTREME
·Brand X Internet
·RoadRunner Cable
·Vonage
| reply to dellsweig
UPnP causes a lot of problems!
I find that in many Linksys units UPnP seems to cause a LOT of problems. I've had several Linksys routers (both wired and wireless) that constantly disconnected clients, crashed, etc. Turning off UPnP in these units seems to make things a LOT more stable! | |
ChrisDAT
Google Keyword Compsysnyc
join:2002-02-26
Hollis, NY
| reply to dellsweig
Port SCAN...
Hello again biker --- The challenge, yes - and, I don't want to take this BEFW11S4 over to my Auntie's house (where she uses Optimum Online) after "burning in" the box for so long, then, as soon as I walk out the door, the router freezes -- then I don't look too good.
To "see" the effect of port filtering on the part of Verizon, I posted on the Verizon board -- Port 80 is filtered and "maybe" 25 [SMTP-mail]. To find out for sure, I used the DSLR port scan applet. I also used the LinkSys LogViewer to see what ports were being tested, and what ports were getting in -- I didn't see any port 80 attempts reach the LinkSys, however port 25 attempts were made [and of course absorbed] by the LS. The port scan also tests UDP ports, but as we know, the LS only logs TCP requests -- so my test says alot, but it's not the whole picture.
So, that narrows it down to a possibly "evil," malformed port 80 TCP [or UDP?] packet that we want to "catch" inside our LAN. Note that not every port 80 packet will be the bad one --- Hmmmmmmmmmmm, I wonder how many people who have DMZ on a PC or an HTTPd on port 80, or whose ISP is VOL have these random but increasingly frequent "lockups?"
Hey 'ster -- I know what you're saying -- that UPnP is "just too natural for me" too -- until I figure out how it can allow any PC on the LAN (wired or wireless) to gain command and control over my LS box without authentication, it stays OFF. | |
big greg
Premium,MVM,Ex-Mod 2005-6
join:2003-10-11
Boston, MA
clubs:
| Good news or bad news?
So here are the results of my testing. All I did was set a separate static address and put little Linky behind a NAT router here. There was other traffic on the network, but no traffic from the outside.
Initial results were inconclusive, as when I moved the wires I rotated it 45° which caused loss of signal.
After I moved it back to its old spot, I power cycled it and and 24 hours of runtime later it hung 4 times, 3 times today between 10 am and 3:30 pm (light usage by 3 laptops).
While there was other traffic from 3 other W2K systems on the NATted segment (NATted segment is on a switch, and one switch port was connected to the WAN port of the router), none of it was the sort of port probing garbage I find on the unwashed Internet.
I could set up a NAT router with *nothing* but the router on the wire, but I suspect it would make any difference. I could also turn on logging, but again I suspect it wouldn't make any difference.
This leads me to believe that the type of connection and the activity on the LAN port have little to do with the hangs.
I have discovered that there are 3 other 802.11b networks in the area (but none of them are on channel 11 with me).
I'm wondering if the hangs are related to the other wireless equipment in the area.
I'm also about to throw this thing down the stairs. | |
biker45
join:2003-10-18
Erie, CO
| reply to dellsweig
Re: Looking for the Common Thread in router problems
big greg: Thanks for your efforts at trying to nail down this problem (random hangs).
I'm with you, I want to throw my linky down the stairs.
Here's a suggest test to eliminate wireless from the equation. What if we disabled wireless on the BEFW11S4 and waited to see if it hangs. If it does, then wouldn't that eliminate stray (or otherwise) wireless traffic from the list of usual suspects?
My guess is that even with wireless disabled, the BEFW11S4 will continue to hang.
Next test: remove the WAN link (cable modem in my case) then wait and see if the BEFW11S4 will hang (thus totally eliminating incoming packets from WAN as a cause of the problem). I'm not sure on how this test might turn out, but as long as I'm predicting, I'll guess that the linky will still hang (thus implying a serious and hard to find flaw in the firmware).
I'll give these two tests a try this weekend (and others may want to try the same, so we can compare results).
On a positive note, I just configured a Cisco Aironet 350 card on my laptop (static WEP) and it works fine with my BEFW11S4. However, there were two paramaters (fragmentation threshold and RTS) that needed to be changed (slightly) on my BEFW11S4. In the case of these two parameters, the max value that the Cisco Aironet 350 allowed was slightly lower than the value I had set in the BEFW11S4 (I had these values set to max allowed by BEFW11S4). So I reduced the values in my BEFW11S4 to match the max values allowed on the Aironet 350 card.
If anyone is interested, I can provide the details. I don't recall the exact numbers off the top of my head, but the reduction was about 10 or 20 (from values around 2000). These changes did not affect the reliability of my BEFW11S4 (it still hangs about once a day). My laptop with Aironet 350 card works fine (as long as the BEFW11S4 remains up). | |
ChrisDAT
Google Keyword Compsysnyc
join:2002-02-26
Hollis, NY
| Re: Wireless Drops...
You may want to examine this thread regarding wireless "drops" every 3-60 mins:
»[wireless] Connection drops every 2-3 minutes
it also includes a WinXP wireless "fix" that addresses a known problem. This may or may not influence the box getting hung up, but who knows? | |
biker45
join:2003-10-18
Erie, CO
| reply to dellsweig
Re: Looking for the Common Thread in router problems
I disabled the wireless feature on my BEFW11S4 v4 (firmware 1.50), and as usual (sigh) it hung within 16 hours.
Like clockwork, my BEFW11S4 v4 has been hanging (randomly) once or twice per 24 hour period (thankfully, I don't have the problem mentioned in the "connection drops every 2 - 3 minutes" thread).
Since my router hung with wireless disabled, I think I can eliminate packets coming in through the wireless interface as a suspect (i.e., interference from my neighbor's wireless networks, either intentional or otherwise).
Next, I will disconnect the ethernet cable between my cable modem and BEFW11S4, and wait to see what happens.
If the router again hangs within 24 hours, then I think I can eliminate any incoming packets as the cause of the hang. Then, it may be time to throw the box down the stairs
I'll post results later.
Anyone else with a BEFW11S4 v4 (firmware 1.50) who would like to perform similar experiments? It would be interesting to see if others who are experiencing these random hangs have similar results. | |
ChrisDAT
Google Keyword Compsysnyc
join:2002-02-26
Hollis, NY
| BEFW11S4 behind firewall/router...
My BEFW11S4v4 w/FW1.50 is behind the router [BEFSR41], and has no wireless clients most of the time -- basically it acts like a switch and WAP, with one wired PC that's on 24/7. This PC has no problem communicating with the other PCs wired to the BEFSR41 or the internet through the BEFSR41. The BEFW11S4 also has an active DHCP to provide IP config info to the wireless clients. All other IPs are static.
I'm intereested in seeing what your setup turns up. | |
biker45
join:2003-10-18
Erie, CO
| reply to dellsweig
Re: Looking for the Common Thread in router problems
Here's the results of my latest test:
I disabled wireless on my BEFW11S4 v4 (firmware 1.50), and then disconnected the ethernet cable between the Linksys box and my cable modem.
The router hung (sometime between 11:33 pm last night and 9:50 am this morning).
So, both the wireless and WAN interfaces have been eliminated as the source of my problem.
To say that the BEFW11S4 v4 (and its firmware) are a POS would only be putting it too kindly.
To attempt to solve this problem, I have previously:
* set my MTU from 1500 to 1400.
* changed from DHCP to static IP addressing
* turned off logging
* made sure that UPnP is disabled
* did a "long" reset
* set the router on 35 mm film containers to improve air flow around it.
It does not appear to me that any amount of configuration tweaking can or will resolve this problem. It's just plain poor quality on Linksys' part.
I will (again) be contacting their tech support, this time to demand that they just give me my money back. I'll escalate to the CEO if necessary (Victor Tsao).
I'm curious if anyone else experiencing the hang problem on their BEFW11S4 v4 has tried disconnecting their cable modem or DSL and disabling wireless (to see if their problem also occurs with both the wireless and WAN interfaces down). | |
big greg
Premium,MVM,Ex-Mod 2005-6
join:2003-10-11
Boston, MA
clubs:
| reply to ChrisDAT
Re: BEFW11S4 behind firewall/router...
said by ChrisDAT :
My BEFW11S4v4 w/FW1.50 is behind the router [BEFSR41], and has no wireless clients most of the time -- basically it acts like a switch and WAP, with one wired PC that's on 24/7. This PC has no problem communicating with the other PCs wired to the BEFSR41 or the internet through the BEFSR41. The BEFW11S4 also has an active DHCP to provide IP config info to the wireless clients. All other IPs are static.
OK I think we eliminated the internet connection as being the problem (cable, DSL, T1, behind router, not behind router, static IP, dynamic IP, PPPoE, and on and on).
I noticed my unit was crowded around a lot of other equipment that could be bothering this plastic-cased device.
I have one room that is wired ethernet, so I put the POSW11S4 in that room by itself. No source of RF or any other electronic device within 15 feet of it. Plus, it's cool... like 60 degrees in that spot.
It hung like a champ.
So it doesn't appear to be heat, or proximity to any RF emissions. And we now know that shutting off the wireless doesn't prevent a hang.
Is Linksys still RMAing V3 units for these V4s? | |
ChrisDAT
Google Keyword Compsysnyc
join:2002-02-26
Hollis, NY
| reply to dellsweig
Re: Looking for the Common Thread in router problems
What we probably need is for someone like Consumer Reports to "get" about 100 of the questionable LinkSys routers, put them in the same room [at "room temperature"], and just plug them into AC and each other and then "wire in" one PC, no internet, no wireless, no nothing. All this PC has to do is let out a broadcast every now and then -- to keep the network "alive."
I would not be surprised if "some" of the little blue devils succomb to this "torture test" before 12 hours. In a week there may be more casualties. After a month....?
You have at least been able to determine big-g, that there must be a hardware problem going on here... The cause may only be known to LinkSys... Maybe the units cannot overheat even once, maybe they cannot tolerate the wrong cable being inserted even once, maybe thay can be damaged by "samsonite" handling on their way to a retailer or to your home, maybe they have cheap flash RAM that "forgets" under certain conditions? -- one thing is certain, the unit you have is bad, and there is probably no flash-ware that can correct the issue. One sure sign is that operating on the switch logic alone appears to be able to crash the unit over time -- a thumbs down for LinkSys.
If you cannot RMA the box, pop the case, and look for stuff that may appear to be burned or melted inside, or soldered connections that may be loose or broken, or anything loose inside the case -- those are quality control issues and signs of internal trauma to the unit. If you decide to just buy another unit, there may be a real solid state electronics whiz that would perform an autopsy to see what is actually wrong with it.
Overall as network hardware goes, they can be considered disposable, since their price is almost equal to the price to ship them for service or replacement. Like I said before, this is the Cisco name goin down the tubes here, so I would expect to see changes [possibly evidenced by price increases] in the near future.
I guess one could say that the hardware is one "thread" that can be tested for and either determined or ruled out as an issue if you are willing to go to the same length as big greg -- I appreciate your effort. | |
|
