I am having a problem that I need your input to help resolve it. I have three servers Ex490 upgraded to a EX 495, EX 470 with upgraded memory and CPU and a WHS 2011 so I am very familiar with building and using these servers. The server I am having the issue with is the EX490. This server has worked perfectly for about 15 months. Suddenly I cannot log into the remote desktop connection or the home server console. I have nine computers and none of them can reach the server.

When I try to access the server using a remote desktop I get a blue screen that looks just like the remote desk top except no program icons appear nor does the log in box to enter your password. I am forced to do a hard reboot and the server comes on and works fine. Sometimes when I cannot log into the server with the remote desktop I can click on the network and access all of the shared files? Sometimes I cannot.

After a hard reboot the server works fine for 2 or three days and I have the same problem identically the same.

Today I could not access the remote desktop nor the home server console. I could see all of the files by opening the network connection. I had my hardware connection device attached and my monitor, keyboard and mouse were already hooked up so I tuned on the monitor. The server was running perfectly and the home console was on and showed no errors. I did look at the logs and here is the only error of magnitude that I could see:
SQLite error

near "=": syntax error

Exception:System.Data.SQLite.SQLiteException

Message:SQLite error

near "=": syntax error

Source:System.Data.SQLite

StackTrace:

at System.Data.SQLite.SQLite3.Prepare(SQLiteConnection cnn, String strSql, SQLiteStatement previous, UInt32 timeoutMS, String& strRemain)

at System.Data.SQLite.SQLiteCommand.BuildNextCommand()

at System.Data.SQLite.SQLiteCommand.GetStatement(Int32 index)

at System.Data.SQLite.SQLiteDataReader.NextResult()

at System.Data.SQLite.SQLiteDataReader..ctor(SQLiteCommand cmd, CommandBehavior behave)

at System.Data.SQLite.SQLiteCommand.ExecuteReader(CommandBehavior behavior)

at System.Data.SQLite.SQLiteCommand.ExecuteDbDataReader(CommandBehavior
behavior)

at
System.Data.Common.DbCommand.System.Data.IDbCommand.ExecuteReader(CommandBehavior
behavior)

at System.Data.Common.DbDataAdapter.FillInternal(DataSet dataset, DataTable[] datatables, Int32 startRecord, Int32 maxRecords, String srcTable, IDbCommand command, CommandBehavior behavior)

at System.Data.Common.DbDataAdapter.Fill(DataTable[] dataTables, Int32 startRecord, Int32 maxRecords, IDbCommand command, CommandBehavior behavior)

at System.Data.Common.DbDataAdapter.Fill(DataTable dataTable)

at HPMediaSmartServer.DataSrvBase.ExecuteQuery(SQLiteCommand cmd)

at HPMediaSmartServer.MediaCollectorDataSrv.GetRequestId(Int64 fileEntryId)

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

+++++++++++++++++++++++

Oddly enough this server works flawlessly and the only issues I ever have seem to be related to forced updates from Microsoft Windows to the server. It doesn’t like to always restart every time.

Just so you know I got so frustrated last week I did a server recover and it worked fine, for 2 days. It is now back to its regular 2 of 3 say disappearing act.

I appreciate an ideas you may have on how to solve this problem.

First thing I would do is install the Home Server SMART Add-In and check the health of the drives. You also don't indicate what other Add-Ins or software may be running on the server.

_________________
Alex Kuretz
Remote Notification: Monitoring and alerting for your Home Server via email, Twitter, text message, RSS, push notifications, and on your Mac desktop
Have you tried our Add-Ins?

Yakuza,
Thanks for the response. I installed Home Server Smart and I have 8 drives installed. I can see the 4 in the server and the four installed in a 5 bay tower with one bay empty and they have blue question marks beside them? I believe this means it cannot read the smart attributes through the eSATA port chip in the Sans Digital tower.

My primary drive is a 1 TB drive and it has an exclamation mark and it shows two status remarks as degraded. They both state reallocated sectors count is degraded but the drive still shows as healthy and failure is not predicted. There are also about 25 other attributes shown for the primary drive and they are all healthy. I also have two messages that state that, “The primary drive has had reallocation events detected and there have been 5 attempted sector reallocation events on the disk. Their contents have been reallocated to the spare area.” Both messages are identical.

I can only assume that I am facing a pending primary drive failure in the near future and the hang ups of the server are being caused by the degraded speed and performance of the hard drive. Would you believe I just purchased a new drive yesterday and it was just delivered?

I have a few questions.
1. At the state that this drive in in is it OK to clone it to the new drive or do a total fresh install which means that I have to rebuild 8 + TB of data.
2. Is it normal for this add-in to not read drives attached to the eSATA port to a Sans Digital TowerRaid TR5M-BP 5 Bay drive tower? I have heard that the eSATA chip in the HP server will not allow the transmission of drive smart data through it from the Sans Digital Tower. Is this why the smart data from those 4 drives is not being read?
3. My drive 0 is a 931 GB drive with only 2 % usage so it probably only has the operating system on the drive. I now notice that under Computer Management the primary Drive C: partition has used 11.36 GIG of 20 GIG of space. The D Drive partition which is 911 GB states that it has 4859.74 GB of free space or 533% of available free space. I didn’t mistype that it what it says. Oddly enough both drives report as healthy?

Thank you for your input and fast response. If you could respond to my three questions then I am ready to rebuild either using the clone method or factory restore or server recovery.

Thank you for your help!

+++++++++++++++++++++++++

I forgot to mention that I use Disk Management 10.0.9.8, Tool Kit 1.1 v 6.0.1800.0, and Grid Junction 1.6 as add-ins.

1. I've cloned a drive with bad sectors, you can see my comment on the process towards the end of the Clone thread in the Software forum.

2. You'll have to talk to the author about that, he checks these forums a lot or you can contact him on his site.

3. This is an issue with the way DE reports storage and is not unusual, search will turn up numerous discussions on this.

_________________
Alex Kuretz
Remote Notification: Monitoring and alerting for your Home Server via email, Twitter, text message, RSS, push notifications, and on your Mac desktop
Have you tried our Add-Ins?

I have an EX490 with a Rosewill RSV enclosure attached and I had the exact same problem you described, and I'm the one who wrote Home Server SMART! So imagine my disappointment when my own software didn't yield results on my own gear.

First item - it sounds like your system disk has bad sectors. Never a good sign. The appearance of bad sectors is regarded as one of the earliest indicators of HDD death. (Bad sectors on an SSD are much less serious.) If you're seeing bad sectors on your system disk, priority one should be looking to replace it. It's a lot easier to replace the disk when you can research and plan for it, rather than when it kicks the bucket and you're left scrambling to fix it.

The ability for HSS to get data is a function of the OS and the device driver. HSS uses only native Windows APIs as its "preferred" approach (it's a lot faster) and, if results don't get returned, then it "falls back" to WMI. WMI, or Windows Management Instrumentation, has two major faults. The first is it's a lot slower than native APIs, and the second is that it doesn't work with USB or FireWire disks at all. The only way to get data from them is native APIs.

When I call a Windows API, Windows takes the call and hands it off to the device driver, which either processes the request and returns the desired data, or returns an error. If an error is returned, the Windows API returns false, meaning the call failed. In which case HSS will attempt to get the information via WMI.

You've probably already noticed that the eSATA port is attached to a Silicon Image controller, and HSS will work when there is a singleton device (single bay enclosure, single disk), but not when a port multiplier is used.

Interestingly, when I upgraded my EX490 to WHS 2011 and installed the latest available Silicon Image driver (for Windows 7, 64-bit), I got a full suite of information for all of the disks in the enclosure. At the time, the native code DLL was still compiled as a 32-bit DLL; re-compiling it to 64-bit also yielded all of the data. This said to me that Silicon Image made changes to their device driver to get the SMART data when a port multiplier is used.

You may thus be tempted to grab the latest available 32-bit driver and test it. If you do, you're doing it at your own risk. I've seen posts all over the forums here about anomalous behavior with some enclosures, and some folks had success in changing the driver, while others took a bad problem and made it worse. So before you perform any driver upgrades, make sure you're not setting yourself up for even bigger issues.

_________________
Matt Sawyer
Owner, Dojo North Software, LLC

HP EX490, 4GB, E5300, 15TB - Server 2012 - Sharing and Streaming
HP EX487, 4GB, E5300, 13TB - Server 2012 Essentials - Backups

I am about to go crazy. This server has worked perfectly for over a year. Last night I spent 7 hours trying to do a system recovery. After 25 boots of the server the red and aqua flashing lights never came on once. I had the two LH lights as solid blue and the RH light as flashing blue. I could read in the server recovery disk but it would either not go in or fail at the 2 or 3rd step.

I had to remove every drive and every cable and only have installed the new disk I was trying to set up. I then tried and finally got a factory reset completed after several tries. Today I set up the server and started transferring in my music only.

After six hours the music was in but the server was unresponsive again. I tried to RDT into the server and I got the same blue RDT screen with no icons and no log in box. I turned on the server monitor and the Home Server Console will not start but everything looks normal and there is no CPU activity. When I tried to turn off the server it tells me that it is ready to install a Microsoft Update. I told it to go ahead and shut down and now I am setting waiting for an hour for it to do something.

I have never seen any server act like this. Why won’t the red and aqua LED flash on reboot? Why does RDT have no log in box? This is the second recovery in a week and I have the same results.

Before I loaded my music the smart add in told me all the drives were healthy and I have already replaced the original bad primary drive. I am lost at this point and frustrated.

I just did a hard reboot and after the server restarted I could RDT into it.

I have several errors in the log and most are related to DOJONORTH source is JoshuaHSS. There are at least 30 of these same errors:

Provider load failure

at System.Management.ManagementException.ThrowWithExtendedInfo(ManagementStatus errorCode)
at System.Management.ManagementObjectCollection.ManagementObjectEnumerator.MoveNext()
at DojoNorthSoftware.HomeServer.HomeServerSMARTService.Enumerator.DiskEnumerator.RefreshDiskInfo(Boolean isManualRefresh, Boolean fallbackToWmi, Boolean advancedSii)
at DojoNorthSoftware.HomeServer.HomeServerSMARTService.Enumerator.DiskEnumerator.RefreshDiskInfo(Boolean fallbackToWmi, Boolean advancedSii)
at DojoNorthSoftware.HomeServer.HomeServerSMARTService.Helper.HssServiceHelper.EnumerateDisksAndSaveData()

For more information, see Help and Support Center at
There are also hundreds if not thousands of information or warnings from source TarynHss:
Disk \\.\PHYSICALDRIVE5 (PNP ID SCSI\DISK&VEN_ST315005&PROD_41AS&REV_CC34\5&2ADEB53A&0&000100) did not return any SMART data. The enclosure bridge chip may not support SMART, or the disk itself is not making the data available.
This I understand as I believe the smart drive data cannot be pulled across the eSATA mux.

The message "provider load failure" is one I've seen returned by WMI on a number of occasions, consistent with a hung computer. Whether the CPU is being consistently maxed out by a process running awry or a hardware problem, this error is returned when WMI components cannot talk to each other.

HSS uses WMI for a number of tasks--things like getting disk capacity info, model number and even the "\\.\PHYSICALDRIVE" path. If the Windows Management Instrumentation service has crashed or is hung waiting on something else, "provider load failure" is a very common error. Many components of WHS, as well as Windows itself, depend on WMI, so if WMI itself (or a component upon which WMI depends) is hung/crashed, you will see problems aplenty.

It is reasonably possible another one of the disks has a problem. Remember that SMART doesn't report everything; SMART is quite imperfect (check out the Wikipedia article on S.M.A.R.T.). I've seen disks with perfect health die without warning, while others with many degraded/failed attributes survive for months or more.

The TarynHss message you showed is indeed the result of not getting data via the eSATA port.

What I would do is try adding your disks back to the server, one at a time. Try letting the server run for awhile with nothing but the system disk, which you indicated is new. If the server locks up after awhile, then you've got another hardware problem. Add a second disk, and let it steep for a few hours. Repeat this process until you get to a disk that gives you trouble.

One other item of note--when I got my Rosewill enclosure, it ran fine for a few days but caused really erratic behavior, including system lock-ups. Turns out the unit came with a shoddy eSATA cable. Replacing the cable fixed the problem.

_________________
Matt Sawyer
Owner, Dojo North Software, LLC

HP EX490, 4GB, E5300, 15TB - Server 2012 - Sharing and Streaming
HP EX487, 4GB, E5300, 13TB - Server 2012 Essentials - Backups

All I can say is (take a deep breath and remember to breath) and really, I don't say this to be an a$$, really, Your data is safe on the other drives, if all else fails you can still access it.
I have been their, you want to hurry and get everything back to the way it was, but ramming and cramming and speed clicking will not help.

It sounds like you chose to do a reset with just the system drive (which is what I would do, recoveries don't work) but even doing a reset, ya need to slow down, during the reset it takes time to write the data and even once your done, it takes time and many reboots to get all the updates. I would reboot and check for windows updates and reboot and check for HP updates and reboot and update till ya get them all. Then I would start adding drives back after checking them all and then lastly after everything is happy, then start coping my data back, knowing it's going to take days and days to get it back to where it was.

All this being said, I will think good thoughts for you. I about threw mine out the window when I found out a recovery was a waste of time, twice.

_________________
[X510 CPU/RAM/All 2 TB Red's] [X510 CPU/Sync'ed Backup/All 2 TB Hitachi's]
[X510 /2 in box]
[X710 LIAN LI DIY W2016E] Stablebit Drivepool [EX-503 LIAN LI 5-bay USB3 External]
Fear the Cloud

I have indeed found another bad drive. It is a Seagate 2TB drive that fails after 5 hours of a long term generic test with bad sectors. CHKDSK says that it found and fixed them but Seagate still says it is bad (checked it twice). I am now checking every drive individually using Seatools on the long term Drive test. This seems to be the best test and the bad drives pass everything else including home server smart.

After testing I will rebuild from zero again with the server having only with four drives and let it run for a few days. After that I will add one drive at a time in the external Sans Digital tower. I want to make sure that I do not have a bad tower eSATA mux. That could also cause the problem I am seeing. That or a bad eSATA cable after using it for 2 ½ years.

Luckily I have three servers as I have had problems in the past and almost everything is backed up and I even have old cloned drives of the primary boot drives of every computer.

My biggest frustration is why the red and aqua LEDs would (will) not flash when I was trying to completely reset the server to a factory reset like it came from the factory. I have done many server recoveries in the past and not getting it to flash like the BIOS says it should just seems wrong!

I have factory reset the server 3 times successfully and it keeps failing. I have installed a new primary boot drive and it has failed twice. I have tested the primary boot drive and installed 3 drives 2 TB in size that I completely checked with Seatools on the long generic test and they all passed. The next day it failed again.

Here it the message I am getting:
++++++++++++++++++++
The previous attempt to open the Windows Home Server Console was not successful because there was a problem with the following file:HomeServerConsoleTab.HPsettings.dll. It the problem persists, please uninstall the add-in associated with this file and try again.
To try to open the console with this file again, click Retry. To open the console without the features associated with this file, click Skip
++++++++++++++++++++++++++++++
I am beginning to think that the original HP disks that I have are defective from HP with a corrupted file. I have not installed the Sans Digital tower so I have the server running with 7TB of drives just to see if it is the server that is having the problem. It is as if the software on the server is defective. I may try to install the HP disks I have from the EX475 server to see if I can get a cleaner install.

After the installation I have updated all the window and HP updates so all the programs are current and I am getting the same error. I logged into the registry and made sure that no Dll files are blocked.

If anyone has any ideas then please let me know or if I solve it I will post the solution here.

By the way the old primary drive that was reported defective with bad sectors with Home Server Smart passed Seatools Long Term Generic test and CHKDSK repair with no issues?

Thanks

Oddly enough I may have found the problem. After four weeks of testing drives and reinstalling the operating system over and over I kept getting the same issues as listed above. I finally came to the conclusion that it had to be one of the add-ins I was using even though I have used them for two years.

I had installed Windows Home Server Disk Management v 1.0.9.8, Grid Junction v 1.5.1, and Windows Home Server Toolkit 1.1 v 6.0.180.0. I removed Windows Home server Disk Management and my server has worked for 5 days now without a problem. Not only is it working without a problem the backups on this server are 3 times faster than ever before.

I must say that I did find a bad drive and I replaced it but even doing so I still had the identical problem after five factory restores. Removing this add-in has greatly improved the speed of the server and in addition it stopped failing every day.

I thought I would post this so others who might experience a similar problem would take nothing for granted. Even an add-in you have used for years may for whatever reason cause a problem in the future.