OK, so removing the port reservation did not seem to have any effect. I was looking at the windowsupdate.log and it appears that most nights (not all, but most) logging stop until the system is rebooted. Below are samples of the final logs until the system is rebooted. Do these mean anything to anyone? Are there any other log files I should look at?

2009-07-27 00:00:10:010 960 274 AU Forced install timer expired for scheduled install
2009-07-27 00:00:10:010 960 274 AU UpdateDownloadProperties: 1 download(s) are still in progress.
2009-07-27 00:00:10:010 960 274 AU WARNING: Failed to change download properties of call, error = 0x80070057

2009-07-28 00:00:10:008 888 87c AU Forced install timer expired for scheduled install
2009-07-28 00:00:10:008 888 87c AU UpdateDownloadProperties: 1 download(s) are still in progress.
2009-07-28 00:00:10:008 888 87c AU WARNING: Failed to change download properties of call, error = 0x80070057
2009-07-28 00:00:10:008 888 87c AU Setting AU scheduled install time to 2009-07-29 04:00:00

2009-08-02 00:00:10:000 892 850 AU Forced install timer expired for scheduled install
2009-08-02 00:00:10:000 892 850 AU UpdateDownloadProperties: 1 download(s) are still in progress.
2009-08-02 00:00:10:000 892 850 AU WARNING: Failed to change download properties of call, error = 0x80070057
2009-08-02 00:00:10:000 892 850 AU Setting AU scheduled install time to 2009-08-03 04:00:00

2009-08-07 00:00:10:000 964 3bc AU Forced install timer expired for scheduled install
2009-08-07 00:00:10:000 964 3bc AU UpdateDownloadProperties: 1 download(s) are still in progress.
2009-08-07 00:00:10:000 964 3bc AU WARNING: Failed to change download properties of call, error = 0x80070057
2009-08-07 00:00:10:000 964 3bc AU Setting AU scheduled install time to 2009-08-08 04:00:00
2009-08-07 00:00:29:765 964 10a0 DnldMgr *********** DnldMgr: Regulation Refresh [Svc: {9482F4B4-E343-43B6-B170-9A65BC822C77}] ***********
2009-08-07 00:00:29:765 964 10a0 DnldMgr Contacting regulation server for 1 updates.
2009-08-07 00:00:30:515 964 10a0 Misc Validating signature for C:\WINDOWS\SoftwareDistribution\WuRedir\9482F4B4-E343-43B6-B170-9A65BC822C77\wuredir.cab:
2009-08-07 00:00:30:515 964 10a0 Misc Microsoft signed: Yes
2009-08-07 00:00:30:515 964 10a0 DnldMgr Regulation server path: http://update.microsoft.com/v6/UpdateRe ... ation.asmx.
2009-08-07 00:00:33:109 964 10a0 DnldMgr Per-Update: d5ea2f40-4dc6-4a9f-917e-17b169221b0d at rate 2500
2009-08-07 00:00:33:109 964 10a0 DnldMgr * Regulation call complete. 0x00000000
2009-08-07 00:00:33:109 964 10a0 DnldMgr *********** DnldMgr: New download job [UpdateId = {D5EA2F40-4DC6-4A9F-917E-17B169221B0D}.101] ***********
2009-08-07 00:00:33:109 964 10a0 DnldMgr Regulation: {9482F4B4-E343-43B6-B170-9A65BC822C77} - Update D5EA2F40-4DC6-4A9F-917E-17B169221B0D is "PerUpdate" regulated and can NOT download. Sequence 4228 vs AcceptRate 2500.
2009-08-07 00:00:33:109 964 10a0 DnldMgr * Update is not allowed to download due to regulation.
2009-08-07 00:00:33:109 964 10a0 DnldMgr Regulation: {9482F4B4-E343-43B6-B170-9A65BC822C77} - Update D5EA2F40-4DC6-4A9F-917E-17B169221B0D is "PerUpdate" regulated and can NOT download. Sequence 4228 vs AcceptRate 2500.

Something I found that helps is actually let the server go to sleep. I don't think these WHS boxes are built for solid 24/7 running. I let mine sleep from 3am to 6am every day and my connectivity has greatly improved with only seldom down times where I have to physically restart the machine.

Sadly after three days of not dying I have to report the problem popped up again.
It even wipes out my secondary USB controller.
This suggest to me it's not a hardware/driver issue.
or only one of the Nics would have stopped.
This is a bit of the OS thats gotten chewed up.
What we know.

1. not all units effected.
2. internal nics as well as external Nics drop connectivity.
3. you must either reset the interface or the system as a whole to regain connection.
4. all the health indicators remain healthy.
5. at least on my system it seems to be limited to evening through very early morning hours. (coincidentally when all the maintenance is scheduled)
6. the few times I have witnessed it through the secondary interface before it too went down, there are no errors/messages.
7. if I manually restart the interface it comes back as if nothing happened.
8. unaffected by traffic load. (Cannot induce with heavy xfers)
9. Not likely heat got my unit up to 55c no problems

Steps taken so far.

1 installed second interface so I can watch the primary fail I have about 1 min before the secondary as well goes bye bye.
2. restart a failed interface fixes it without a total reboot.
3. tried to watch processes havent seen any trends yet.
4. use fixed IP
5. rebuilt TCP stack to original install settings
6. latest drivers for everything
7 for purpose of test disabled the following services
WinHTTPautoProxy
Firewall
Backup
UPNP
Media Connect

Will keep you posted

BTW, this is also being reported under other Windows Server products.
the diff they have multiple redundancy and full time monitoring.

_________________
Current DIY server DL380 G6 2xquad core 2.8ghz 32GB rdimm ram 512mb 2xHP 410 raid with 2x sas expanders (2 X norco RPC-4220) 48 slots stacked with room to grow.
30TB (Movie/File/Music)Win 2012 essentials 64bit Storage spaces
Ex490 E6400 4gb ram

Not an absolute fix but at least you can get back into your box without a hard boot.

You will need the file Sleep.exe from the windows resource kit or just google it.
Place it in your c:\windows directory.
The open notepad and copy the following lines into it.

------------------------------------------------------------------------
:start
@echo off
CLS
@echo %Date% %Time%
ping "Your routers IP Address" -n 1 > c:\ping.log
if %errorlevel% == 1 goto restart
sleep 120
goto start
:restart
@echo %Date% %Time% >> c:\restart.log
netsh interface set interface "Local Area Connection" disabled
Sleep 15
netsh interface set interface "Local Area Connection" enabled
sleep 15
goto start
:end

---------------------------------------------------------------

save this to your c:\ as nicreset.cmd

You need to create an empty text file called restart.log in the c:\ so when your system runs this file you have a log with date/time of every event so you can look into the event logs for issues.

You can lower the 120 sleep time but I would caution against lowering it below 10sec, also I would not change the 15sec on the nic reset below 10sec as this may not give the OS time to sense the state change.

If you have multiple nics make sure to name them correctly.
i.e "local Area connection" or "Local Area Connection 2"
If you miss this part in the re-enable line you can be in for a long night.

you can either run it manually or place it in your startup folder.
it will ping your router every 2 minutes (or whatever you set above)with a single packet and
only reset the network interface if it does not get a responce.

Perhaps saving your system a hard reboot, I have seen it restart the Nic without interupting copies only pausing them but I can't garantee that for everyone.

Be sure to test it by unplugging the network wire for whatever time you set as your test time, after plugging it back in check the restart log for the entry.

Be carefule of typos or missing lines you can disable all of your network interfaces an create a problem.
Use this at your own risk.

I hope this helps.

_________________
Current DIY server DL380 G6 2xquad core 2.8ghz 32GB rdimm ram 512mb 2xHP 410 raid with 2x sas expanders (2 X norco RPC-4220) 48 slots stacked with room to grow.
30TB (Movie/File/Music)Win 2012 essentials 64bit Storage spaces
Ex490 E6400 4gb ram

The claims of a 550mbs to 650mbs throughput cieling are inaccurate.
I have actually hit 95% approx 900+ mbs utilization with a single system doing a file copy.
during my testing trying to induce a failure.

using the 03/03/08 driver..

I have had the above script running for a day now with no breaks.
very little to no cpu usage at 20sec intervals.

_________________
Current DIY server DL380 G6 2xquad core 2.8ghz 32GB rdimm ram 512mb 2xHP 410 raid with 2x sas expanders (2 X norco RPC-4220) 48 slots stacked with room to grow.
30TB (Movie/File/Music)Win 2012 essentials 64bit Storage spaces
Ex490 E6400 4gb ram

OK, so I now think that I have found the issue, but am still having problems figuring out how to resolve. I was searching through all the log files I could find and ran across the chkdsk_D_.log file. It appears that they scheduled chkdsk at midnight is causing the system to hang. In reviewing the log, it seems to get stuck at around 12% and never progresses. I tried doing a chkdsk manually and got the same result. I then found a few posts on this and other forums that had similar issues. There is a MS hotfix that claims to fix this issue. Installed that last night but still the system was hung this morning. Looking at the log, there is now some corruption in the file. (See below). Before I call HP and have to either RMA or reformat this drive, are there any suggestions out there on other things I can try?

snip of log........
10 percent complete. (5000 of 64448 reparse records processed)
10 percent complete. (10000 of 64448 reparse records processed)
10 percent complete. (15000 of 64448 reparse records processed)
10 percent complete. (20000 of 64448 reparse records processed)
10 percent complete. (25000 of 64448 reparse records processed)
10 percent complete. (30000 of 64448 reparse records processed)
11 percent complete. (30946 of 64448 reparse records processed)
11 percent complete. (35946 of 64448 reparse records processed)
11 percent complete. (40946 of 64448 reparse records processed)
11 percent complete. (45946 of 64448 reparse records processed)
11 percent complete. (50946 of 64448 reparse records processed)
11 percent complete. (55946 of 64448 reparse records processed)
11 percent complete. (60946 of 64448 reparse records processed)
12 percent complete. (61990 of 64448 reparse records processed)
126444864448
64448 reparse records processed.

CHKDSK is verifying indexes (stage 2 of 3)...
12 percent complete. (1 of 619294 index entries processed)
3@
o@çÇ
çòªA dbwÿ. 1x­?i`·‰B"cQ_ ñÿy™/YS eo .tÿ!ò!?
7€Õ/cS Rfàw $aG·?v±$¿t @;5L 5 t S w`um¿#—²#½ÏCà!n £Ïþt !½¿‡)ñ#¿
ãäû?7çäM u«ì·qïÿÿ™¿ôýŸ¿ƒ¿ƒ?±ƒñ-_z?-Ý(_ [ ׿ rl0,·Ï?v?:WD¿[ ÷H_CTð>_zŸ]10þ ¿6°•??_
_
W
¯6ÿ÷[™iò™ÿy~_-ßbµ§?vñ©ó§ _z³_zü
_z_z_z_zÿ)ÿ)ú[p¹nPÑ__z»ÿ1¿ßHu_vßÕùÿ[ ÿÿßÕŸ,7_zÿÿnðbq[ ÿ

ÿuV›må_[°uo ¿~..ý1PlºLv¿ßp[Ù?ï?9_ÿt ¿??¿_W__Ÿ
Ÿ
i°Uÿ•_v? ? 7 ÿÿLõÿ¿ô¿ô?Bÿ^¿ôù»cí² b u t e  R Pl i  a b l X[ˆ€c o r d <‚d ŒT V ] = 0

jA zt Bi j’W M C nn nÝ nc B7–•P ±
¡¸i c4‹+A@Y€‘¢u€mc h€os€…êT€m€
r€
³½)RV€åÛo s´“­Q1V}O PÀGo@Gw QC C O@™:µÀ^C@ŸeÀ]C±iÀeªsÀaÂgeB+Q@ŒŠe@ yÀb jĤ¬s AA-tÀrÀEÁg@1 3@
o¯@´Ç
ÕºO=A@dÂ$®wô]kO BB8cö¢·ÿ¹†/YS uá@tÿ!f q‡?
·‘ï#SÕ`“f`Dw $r`%?v/±¿í¢/L t èS w`um¿#²#Vr 96Cbµv`rÕ "e ½ à¼i¢¾wgaÇñ#¿
ûÈï%Mí us ?¿ r%ïÿ™%çáOàâi ã$ ÈL y }n ¿G±#ß sëo"!¿ôt ;ë@Ÿ±¿¿ƒ[ ?v?:7)¿¿"9‡‘lÿ!×_C_z? ? YŒ?ÿ??7_
?vÿ? 9 ù _-_žUž?v¿_z_zcpGÑu_z‘£_z_z_zŸWeÿ²;×!?v__YÿŸ_v?lµµÿÿÿ_UŸ,_zÿÿ÷ÿ

ÿuÿbMSgÿW __[Ÿv¿HÿÏŠoÒae°t [ÐbþeXŸ~¥2_“
_zï?¾?3_V[tÚ__ÿ_Ÿ
sã_vŸ$7ÿÿÿL#u#ŸuQ;_z=Bßú[ðÂl0eÓ_
_ ÿU ñ_z_ _ ¿ô¿_z_z_z1vø² u b l i €c ] = 0 0

A Št r Èb u $e dW M C (o n e Ät [ P :'‚R eÚaêe ~Y aU An Mh s eTõ m r 3“=SRŠV€ƒd€Oo s´“‘ 1–´@P h€¼t€
µ@‘ ƒ@C OE@v:À^C r@9aUÄ{iÀysÀaÀ
c«ÀÁ…Q@ke@ yÀ(b j@c@s kAA-
.
.
.
.
.
.
.,H#R‚Ô€õ
àÈ 8² Ž,H}R Ô€#
àÈ‚8² õ,H} R Ž€#àÈ‚Ô² õ
H} 8 Ž,€#RÈ‚Ô õ
à} 8²
Ž,H€zÄ#R‚Ô€õ
àÈ 8² Ž,H}R Ô€#
àÈ‚8² õ,H} R Ž€#àÈ‚Ô² õ
H} 8 Ž,€#RÈ‚Ô õ
à} 8² Ž,H#R‚Ô€õ
àÈ 8² Ž,H}R Ø€#
àÈ‚8² õ,H} a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"è
Ža"€„ Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"è
Ža"€Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"è
Ža"€˜ Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž Žtdtda"a" Ž
index $R of file 26.
Inserting an index entry into index $R of file 26.
Inserting an index entry into index $R of file 26.
Inserting an index entry into index $R of file 26.
Inserting an index entry into index $R of file 26.
Inserting an index entry into index $R of file 26.
Inserting an index entry into index $R of file 26.
...........

Did anyone have success here? What was the eventual fix?

I am having this issue ever since moving to a new house where I now have linksys router and comcast.

Yesterday, I lost contact with the MediaSmart and had to hard boot. This morning, the same condition exists. Totally unable to communicate using the console app or RDP.

Any suggestions on how to communicate to resolve other than hard booting - which doesn't seem to work anymore? Hard booting brings it back online, but not reachable.

_________________
MSS (7.5TB Total): HP EX495, 1x1.5TB, 3x2TB (Seagate and 3 x WD Caviar Green), 2GB RAM, PP2
STORAGE (7.5TB Total): Sans Digital TR5M-B Ext. Chassis, 5x1.5TB WD Caviar Green
UPS (1500VA, 865W): APCC Back-UPS XS 1500 LCD w/GridJunction

I fixed my exact same problem. It was TwonkyMedia Server for HP Media Smart.

I uninstalled Twonky and the box hasn't needed to be rebooted since.

If ONLY I could get the server to respond to RDP or the Console App.....

_________________
MSS (7.5TB Total): HP EX495, 1x1.5TB, 3x2TB (Seagate and 3 x WD Caviar Green), 2GB RAM, PP2
STORAGE (7.5TB Total): Sans Digital TR5M-B Ext. Chassis, 5x1.5TB WD Caviar Green
UPS (1500VA, 865W): APCC Back-UPS XS 1500 LCD w/GridJunction

Do a hard reboot via button on back

I have dne that and was not able to get communication. Yesterday, I once again hard booted and quickly did an RDP into the server to discover that an NTFS file corruption issue was the problem. I had a CHKDSK run all night and this morning things are fine. Thank Heaven!

_________________
MSS (7.5TB Total): HP EX495, 1x1.5TB, 3x2TB (Seagate and 3 x WD Caviar Green), 2GB RAM, PP2
STORAGE (7.5TB Total): Sans Digital TR5M-B Ext. Chassis, 5x1.5TB WD Caviar Green
UPS (1500VA, 865W): APCC Back-UPS XS 1500 LCD w/GridJunction