Dell iDRAC 6 Virtual console connection error

I am system administrator for national Scouting organizator in my country and a few days ago I had to preform some tasks on our two Dell R510 servers that are equipped with Integrated Dell Remote Access Controller 6 – enterprise version (iDRAC6).  After connecting over Cisco VPN, I could access the iDRAC web application, but after launching Virtual console, connection returned error: “Connection failed.“.



I already had problems with Javas security permissions and iDRAC virtual console, so firstly I went to Java Security settings and  I changed settings to allow everything – from false certificates to unknown sources and everything in-between. After retrying I still got error and enabling console gave me following error:

After trying literally everything (changing virtual console settings in iDRAC, resetting iDRAC, updating iDRAC to latest version, even changing registry as suggested by: [1] and [2] ) i stil could not connect. I don’t remember if registry changes solved that java.util.prefs.WindowsPreferences error, but nevertheless I still could not connect to virtual console.  Last option coming to my mind was to downgrade Java from 8 to 7. In short this helped.

I have latest iDRAC updates installed (version 2.85 ) released in May 2016 – way later than last public Java 7 release on (April 14, 2015).

TL;DR; Downgrade Java from 8 to Java 7 – build 79 and your problems are solved if you encounter this.


Asus UX31A Heavy Load Temperatures

In the post about how to upgrade SSD drive on UX31A user commenter Iosif Bancioiu asked how  my UX31A’s temperatures are under heavy load and in idle.

I performed heavy load tests in room that had about 23°C, I run stress test with program called HeavyLoad that during test performs:

  • Full CPU load
  • writing to a temp file (ssd stress)
  • allocates full virtual and physical memory (memory stress)
  • simulates disk accesses (more useable in with hard disk, than ssd)
  • stresses GPU with 3D rendering graphics.

I first run the test for 5 minutes to warm up the PC and then I run 10 minute test while I measured temperatures and other parameters with program HWiNFO64.

Temperature results on test are:

  • CPU:
    • min: 88.0°C
    • max: 91.0°C
  • GPU cores (CPU GT Cores):
    • min: 86.0°C
    • max: 89.0°C
  • Intel PCH:
    • min: 70.0°C
    • max: 76.6°C
  • Crucial SSD:
    • min: 53.0°C
    • max: 58.0°C
Screenshot of the test results. Click on the image for full image, with all the details about the test.

If we check Intel’s specifications on i7-3517U it is declared that processor has T_JUNCTION = 105°C. Intel declares: Junction Temperature is the maximum temperature allowed at the processor die. It seems that processor is running as expected and in not too hot.

Regarding SSD, I didn’t found specs for my SSD exactly, but for similar SSDs Crucial is reporting operating temperatures from 0°C to 70°C, which also shows that SSD isn’t too hot.

Windows 8.1 + SFTP/FTPS = security risk?!

During my daily work I often connect to my Linux server, usually that is over SSH connection. But when I want to do some file-transferring I often use SFTP (that is FTP over SSH). My primary OS is around 50% windows 8.1 and 50% Ubuntu GNOME. Today I accidentally tried to connect to address sftp://my.server.tld/ on my windows machine to upload some files. And I have found that SFTP protocol automatically silently (!) fails back to FTP.


Let’s go back, so I wrote sftp://my.server.tld/ into explorer’s bar. After a few seconds windows greeted me with unfamiliar window that asks for username and password. At that moment I realized that I was not connecting from Ubuntu machine and I become suspicious – from when did Microsoft implement SFTP into windows?


In that second I run Wireshark and started to capture traffic and analyses traffic between my PC and my server. I have confirmed that sftp protocol failedback silently to ftp (wireshark logs further down page). The only record that this happened is in login window. This is total security risk as we all know that 90% of users won’t read everything on system login form like this. Just remember what are notifications in any browser if something is wrong with HTTPS’ certificates. Let’s see two cases where this would have gone through without user knowledge (besides the one that small note in the login window is missed).

Case 1: User has somehow saved username and password for this domain and login window doesn’t apear. User would be using FTP without knowing, while expecting that their paswword and files aren traveling in plain text over ethenet.

Case 2: If server would be configured to allow anonymous logins, I would be logged in anonymous FTP. I wouldn’t miss login form, because I would think that I was logged in SFTP via public/private key system. (this one is more obscure, but can also happen – why would you have anonymous login allowed?)

In any case user wouldn’t have known that SFTP connection failedback to FTP. I think this is severe security risk and it should be fixed. It is bad enough that windows ignores SSH and SFTP, but that secure connection fails back to unencrypted in such a suptile way is hundred-times worst.


Same as written above holds for FTPS connections.

Upgrading to 512GB SSD on Asus Zenbook UX31A (update: UX31E too)

In June 2012 I have bought Asus Zenbook UX31A (model UX31A-R4003X). It has ADATA XM11 256GB SSD in similar form type as M.2 – Be careful disk inside UX31A is NOT M.2, they are not compatible. Disk is some kind of proprietary form type only found in Zenbooks. Image bellow is showing my ADATA and M.2 disk.


In 2.5 years I outgrow this disk and I had to upgrade to 512GB. Asus’s decision to use proprietary connector gives us a lot of troubles. To solve this issue we have to use an adapter. I bought this one (see edit 11). Here is an image if amazon stops selling it:


Disk I bought is Crucial M550 512GB SATA M.2 Type 2280 (CT512M550SSD4). You have to be careful because not all M.2 drives will work. It has to support SATA data transfer and not only newer and faster PCI-E mode!

To open UX31A you have to unscrew 10 T5 screws on the bottom plate. To change disks, you have to unscrew additionally one philips-head screew that holds disk in place. Here is a photo of disk and adapter fitted inside laptop. Adapter and disk is a bit thicker in comparison to old one. Bottom lid is a bit harder to put on, because of thickness difference.


To be on safe side I insulated the aluminum lid where it comes in contact with the disk with antistatic bag in which electronics usually comes in.

And because last post was on disk speeds, I must show you results of speed test on new ssd disk (left ADATA MX11 256gb, right CRUCIAL CT512M550SSD4):

ux31a - adata mx11 vs CRUCIAL

As you can see disk speeds were upgraded considerably. We will see how that will transfer to real life.

EDIT: Someone in the comment selection asked for disk speeds of the Transcendent disk. Because CrystalDiskMark changed its test I have redone tests on my CRUCIAL disk, so they can be simply compared.  As we can see Crucial has better writing speeds in nearly all the cases, and similar reading speeds.


EDIT: Check the comments section for additional details (how is disk working out after few days of usage) and tips on how to copy/transfer windows. 

EDIT2: Commentators Uli and Holger report that Sandisk X300 works too. (Thanks Uli & Holger for the feedback! )

EDIT3: Commentators Lars and Werner report that they used Transcend 512G MTS800 M.2 SSD and it works as expected.  Florian also used Transcend’s disk. (Thanks Lars & Werner & Florian  for the feedback! ).

EDIT4: Commentator Stefan used Sandisk X300 SSD M.2 2280 256GBSD7SN65-256G-1122 (Thanks Stefan for the feedback! )

EDIT5: Commentator Stephane used same adapter and “Transcend TS256GMTS800 SSD interne 256 Go M.2 (Type 2280) SATA III” disk on older model – UX31E .   (Thanks Stephane for the feedback! )

EDIT6: I have succesfully installed Transcend MTS400 M.2 SSD 256GB SATA III, MLC into my GF’s UX31A. This disk short one (42.0mm long, Type 2242) and it fits without a hitch. 


EDIT7: Commentator Eric used same adapter and “Samsung EVO 850, 500GB”. He reports some problems with back light and touch pad, we are still trying identify the problems and figure out if they are caused by the new disk  (solved: there were wrong screws in wrong holes).  Jakob reports that Samsung EVO 850, 500GB works but has ” a little tight fit”. Check below for in the comments.  Pete is reporting that Samsung 850 EVO M.2 250GB works as expected. Oliver used Samsung 850 EVO 250 GB M.2 3.5-Inch SSD (MZ-N5E250BW)  (Thanks Eric , Jakob, Pete, Oliver for the feedback! )

EDIT8: Alte has found new type of adapter that should minimize problems with thickness. (Thanks Alte!)

EDIT9: Acording to Michael Crucial 500GB MX200 Works as expected (Thanks Michsel!)

EDIT10: Svankius used Adata Premier Pro series ssd (ASP900NS38-256GM-C) (Thanks Svanikus!)

EDIT11: m3mp5  used different type of the adapter it’s smaller and could be better. (Tanks for the feedback m3mp5!)

EDIT12: Thanks to Sarah, please consider this comment  when deciding about adapter.

Original disk inside UX31A:



Insulating the lid where it could touch the disk with a simple plastic anti-static bag and some electrical insulation tape:



Comparison of disk speeds

When I was buying a new portable drive I always wondered:

How are USB 3.0 disk drive speeds compared to normal drives in my pc?

One day I bought a new  USB 3.0 WD Passport drive and I wondered what are speeds I get from new drive and from my other disks. I konw some of these components are a few years old, but I thought these data could give an idea to users what speeds can they expect.

These are some disks that are in my pc/laptop and will be compared bellow:

  • Intel X25-M SSD (SSDSA2M080G2GC) – connected to GA-P55-UD6 motherboard
  • RAID0  2x 1TB Segate Barracuda 7200.12 – connected to RAID controller on GA-P55-UD6
  • Seagate Barracuda 7200.11 connected to GA-P55-UD6
  • WD Caviar 2 TB (5400 rpm) connected via eSata to GA-P55-UD6
  • ADATA XM11 ssd inside ASUS UX31A
  • WD My Passport Ultra 2 TB connected via USB 2.0
  • WD My Passport Ultra 2 TB connected via USB 3.0 to cheap PCI-E ST-Lab USB 3.0 card
  • WD My Passport Ultra 2 TB connected via USB 3.0 to Asus UX31A

For comparison I used Cristal DiskMark 3.0.3 x64. Results for sequential read/write (in MB/s):


Results in 4K random read (in MB/s):



We can see that the price (one tested is cheap, one is inside Asus’ top of the line ultrabook) of usb 3.0 controller will not noticeable effect on speed of external hard drives.

You can expect 3x the performance on USB 3.0 in comparison to USB 2.0 in sequential operations, but no better performance if your operation is mostly random reading.

For sequential operations there is no difference between intermediate SSD drive and two drives in RAID0, but there is huge difference for random reads.

Original screenshots if you are interested:

Intel, Seagate, WD Caviar:

JurePcADATA inside ASUS UX31A

WD Passport:


I2C addressing on STM32F4-Discovery

In past month STM published new set of tools and software examples that implement HAL (wikipedia: Hardware abstraction layer) on the many of theirs µC. They are calling this set of tools STM32CubeF4. This post is a little public note to myself and all others who are banging or will be banging their heads at wall because they aren’t able to make I2C communication to work.


I2C addressing (not the 10bit, i am talking about 7bit) has only 7 bits because one bit of 8 bit addressing space is reserved for flag denoting if operation will be reading or writing. That one bit is LSB (Least significant bit). If you are into details, you can more here.


Till now i worked with libraries that input a 7bit address and they automatically shift it for one bit to left. After that libs set LSB to the right value depending on the operation.

 STM32CubeF4’s way

(At the time writting this: HAL version is 1.1.0 )

Long story short: You have to shift the 7 bit address yourself. You can see two functions from my I2C library that are more or less self explanatory – note the shifted addresses  (uint16_t)I2C_ADDRESS<<1 


And ReadBuffer :


I hope i saved you an hour or two :-).