CogX's Ramblings

Silverlight and Firefox

2011-05-09T17:30:00.004-05:00

On my primary home Windows PC, somewhere along the way, the Microsoft Silverlight plugin stopped functioning with Mozilla Firefox. True, there had been several updates to both Silverlight and Firefox on this computer, since the last time I recall that it worked, but it still worked on my other PCs. What was different on this one computer?

I was stuck in a loop, where a web site, such as Xbox.com, would tell me to "upgrade to the latest version of Silverlight" in order to view their home page video and yet the Silverlight download page told me it was already installed. On the Netflix.com web site, when I tried to play a movie, it just took me to a page that said "Thanks for installing". In other words, Silverlight was properly installed and yet it still didn't work with Firefox!

Sad to say, it took me way longer than it should have to figure out the problem. Back at the end of August 2010, Microsoft came out with update KB2264107. What I had done on this particular computer was to set the Session Manager key CWDIllegalInDllSearch value to 0xFFFFFFFF. On my other computers, I had set the CWDIllegalInDllSearch value to 2.

One solution would be to change the CWDIllegalInDllSearch value under the Session Manager key, but I wanted to keep it as it was. Instead, I added "CWDIllegalInDllSearch"=dword:2 under the HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Image File Execution Options\plugin-container.exe key I created.

Actually, first I added a firefox.exe key, but when that didn't work, I immediately recalled the change Mozilla made back in v3.6.4 adding crash protection for third-party plugins. If dom.ipc.plugins.enabled.npctrl.dll is set to False (or dom.ipc.plugins.enabled is set to False and dom.ipc.plugins.enabled.npctrl.dll is not set to True), then one would need to use the firefox.exe key or else enable crash protection for the Silverlight plugin (npctrl.dll).

Seagate HDD "thumping" noise (or not ?)

2010-06-05T18:50:00.001-05:00

Update: It would seem that the problem is not these Seagate HDDs after all, but the Synology DS710+ NAS itself is to blame. I put a spare Western Digital WD4000AAKS drive in the DS710+ to see how it would sound, after pulling those two Seagate ST32000644NS drives out, and it still periodically makes that same thumping noise. The problem appears to be the combination of the plastic HDD rails used in the DS710+ and how this enclosure acts as an echo chamber to amplify the HDD noise.
Here's what that single WD HDD sounds like (with background noise reduction) and what the waveform looks like.

I bought a Synology DS710+ NAS last week. It ships without HDDs, so I purchased two Seagate Constellation ES SATA 3Gb/s 2-TB Hard Drives (ST32000644NS) separately. This particular HDD model is on Synology's extensive drive compatibility web page.The main reason I purchased it was I wanted a 2TB SATA drive that was quiet. Tom's Hardware - Benchmark Sound Level chart shows this Seagate Constellation ES model as one of the most quiet.

Well, unfortunately, although it is mostly quiet, it periodically makes what I consider a "thumping" sound. These drives passed the extended S.M.A.R.T testing and the very long NAS volume creation. However, this thumping sound really stands out, much more so than the usual sound of active hard disk drives. I've been around multiple PCs and file servers now for over 17 years, so I absolutely know what is typical for hard disk drive noise and these Seagate drives are making a racket I just don't expect to hear every few seconds.

Of course, only after purchasing these new Seagate HDDs, I searched and found a reference to the same sort of thing for an older Seagate 750MB (7200.10) HDD reviewed over at AnandTech in May 2006:

The drive is basically a "silent" component in our case during idle conditions with only the occasional disc whirling but at full load the drive does generate a noticeable, almost thumping sound. In fact, although two of the drives tested are considered louder by our test equipment, the noise tone was slightly heavier which elicited our immediate attention to it.

I used a Nokia N900 held close to the back of the NAS to record the sound. I did the same next to the front grill of my Dell PC which has two internal HDDs, a Samsung HD502IJ and a Western Digital WD4000AAKS.

Audio clip 1a: Two Seagate ST32000644NS HDDs in a Synology DS710+ at idle (no connection made).

Audio clip 1b: Audio clip 1a with 70% background noise reduction (Adobe Audition 3.0).

Audio clip 2a: Two HDDs in a Dell PC while doing a large file copy between the two drives.

Audio clip 2b: Audio clip 2a with 70% background noise reduction (Adobe Audition 3.0).

The sound clips with the background noise reduction are easier to listen to and hear just the hard disk drive noise. In clip 1b, notice the spike in sound at 2 seconds in and again at 37 seconds. When looking at the clip 1b waveform, from Adobe Audition, it is easy to see how much louder these "thumps" are than anything else. These spikes are just as easily seen in the clip 1a waveform, as well, which is the whole point that they are that much louder than the background noise (including the NAS enclosure's cooling fan).

Now, between 3.7 seconds and 4.7 seconds into clip 1b, the Seagate HDDs can be heard making noise again, but not nearly as loudly. The clip 1a waveform shows it is just a bit above the background noise, although it is still quite noticeable in the clip 1b waveform. The same sort of thing happens between 17 and 18.3 seconds and again between 24.4 and 25 seconds. At 42 seconds and until the end of the clips, the noise is louder, although not quite as loud as at 2 and 37 seconds, so sort of a "mini-thump".

Compare the Seagate HDD noise to the two HDDs in my Dell PC. True, the Dell PC case probably does dampen the sound a little bit more than the Synology NAS, but this being a BTX style chassis there is a grill in the front where I recorded clip 2a/2b and I can always hear the HDDs when sitting in my chair using the computer. Like I said earlier, after 17 years, I'm used to hearing hard disk drives working away, as they are mechanical devices after all. The clip 2b waveform does show where one can hear some HDD chatter between 2.8 and 4.1 seconds, as I was copying a large file between the two drives at the time of the recording (whereas I haven't even starting accessing the NAS volume yet to make them do any work). However, it is definitely nothing that jumps out at you and when one looks at the clip 2a waveform it is harder to quickly pick out the HDDs in the overall background noise.

To sum up, there is no perfectly quiet hard disk drive and in a small NAS enclosure I was even expecting to hear them a bit more than the two drives in my PC, but this periodic thumping sound of the Seagate Constellation drives is just not what I was expecting.

To make a long post even longer, I've had a Netgear ReadyNAS at work in my office since December 2009, on top of a table right next to where I sit. It has four, not just two, Seagate Barracuda ES.2 SATA 3.0-Gb/s 1-TB Hard Drives (ST31000340NS) and oddly enough the four of them together are not as loud as just these two Constellation ES drives here at home. I have heard some "mini-thumping" every now and again at work, but nothing that made me frown in disgust like I did after only a few minutes of powering up my Synology NAS at home for the first time yesterday. I suppose I'm also not as picky about annoying sounds at work (sort of the baseline) as I am at home.

Comedy: Privacy? (Pete Holmes)

2010-03-20T15:16:00.001-05:00

I'll never understand why so many people are willing to post all details of their lives on the web!
Pete Holmes - Privacy Is Uncool

Comedy: Passwords! (John Ramsey)

2010-03-20T15:06:00.002-05:00

Passwords are utterly annoying, but that's why I find this bit so funny!
John Ramsey - Passwords

Comedy: Texting (Paul F. Tompkins)

2010-03-20T14:59:00.005-05:00

One of my favorite stand-up comedy bits by Paul F. Tompkins...
Paul F. Tompkins - Phone Lover

Sentinel LM Files and AntiMalware False Positives

2009-12-26T16:16:00.001-06:00

Since 1994, I've been installing SPSS for Windows (renamed to PASW Statistics, starting with v17). One of the issues that comes up from time to time is how the Sentinel Licensing Management (LM) product (currently owned by SafeNet, Inc., formerly by Rainbow Technologies, Inc.), which SPSS utilizes for software license enforcement, creates very small (sometimes 0 byte) files. These files have .dll and/or .tgz extensions and some appear to only be created during installation, while others, if removed, are created every time the SPSS/PASW application is launched.

Here are the most commonly created Sentinel LM-related file names I have found on our systems, created by installing or running SPSS/PASW versions 18, 15, and/or 14.

SPSS v14: nsprs.dll, nsprs.tgz, ssprs.dll, ssprs.tgz
SPSS v15: lsprst7.tgz, lsprst7.dll, sysprs7.tgz, sysprs7.dll, tmpPrst.tgz
SPSS v18: grcauth1.dll, grcauth2.dll, prsgrc.dll, prsgrc.tgz

Here are the names of files found under the %windir%\system32 folder which, at least in my experience, are only created when first installing SPSS/PASW :
clauth1.dll. clauth2.dll, serauth1.dll, serauth2.dll, and servdat.slm (a hidden file)

I have found that deleting these files does not prevent SPSS/PASW from launching and running properly.

The primary reason for this post is that I've found some AntiMalware scanners report some of these files with the .dll extension as being malicious. For example, Malwarebytes logs serauth1.dll and serauth2.dll as a "Trojan.Agent".

For example, do a web search and the serauth1.dll and serauth2.dll file names often come up with people asking what to do about their Windows PCs when a scanner detects those files as being malicious. Now, of course, any real malware can use those file names if they so choose, but I have found several different forum board posts where the information the user includes clearly shows that they have either SPSS/PASW or CGTech VERICUT installed and both of those applications are known to utilize the Sentinel LM components.

Full AV software products will rarely, if ever, flag actual Sentinel LM-created files as malware. In fact, I have uploaded several of these files to Virustotal and none have ever been flagged by any of the AV products they test files against. Especially when the files are 0 bytes in size, as is the case for serauth1.dll and serauth2.dll (when SPSS/PASW creates them anyway), clearly some AntiMalware scanners are flagging some of these files simply because of the filename and not because of any actual executable code (malicious or otherwise).

Several of the binaries included with an SPSS/PASW installation have references to the files above, such as: lsapiw32.dll

Version info, as included with SPSS v15:
Rainbow Technologies, Inc.
LSAPIW32
7, 3, 0, 6
Integrated Client DLL
Copyright ? 2004 Rainbow Technologies, Inc.
lssrv32.dll
Sentinel LM
7, 3, 0, 6

Version info, as included with PASW v18:
SafeNet, Inc.
LSAPIW32
8, 2, 2, 300
Integrated Client DLL
Copyright (C) 2008 SafeNet, Inc.
lssrv32.dll
Sentinel RMS Development Kit
8, 2, 2, 300