Klystron, Kernel Update a Success

The laptop which I name ‘Klystron’, received a Kernel Update on July 7. This brought its kernel version to 4.4.0-30-generic as far as I can tell.

Prior to this update, one of the problems the laptop was still experiencing, was due to its Realtek WiFi chip-set, and the kernel module ‘RTL8723BE’. It had a malfunction in how it was connecting to my WiFi, which should have set in twice again, since July 7, according to earlier observations I had made. This bug had become quite predictable.

But since the latest kernel update, this malfunction has not been taking place anymore.

So I would say, that this kernel update was a huge, welcome success.

A separate question I have not yet answered, was whether its Suspend Behavior has also been corrected. This will be slightly more complicated for me to test, because as the above posting suggests, I had adapted a script, which seems to correct it. If the problem had been resolved in the kernel update, my script would simply not do anything. And the end result would remain, that the problem is not apparent.

In order to see whether the kernel update actually resolved that issue, I would first need to disable my script, and then try a few Suspend Cycles, since this problem was also not taking place with 100% consistency.

I have not yet committed myself to doing that.



Klystron Kernel Update

My Linux laptop named ‘Klystron‘ is still fully subscribed to the “Kanotix” repositories. As the reader may recall, Kanotix is a slightly customized version of Debian Linux, that is KDE-based, and that is maintained by a group of developer-experts who I trust implicitly.

Being subscribed to their specific repositories and configuration details has as one advantage, that periodic kernel updates are fed to me, via package manager.

As I came home from camping yesterday, on July 7, I also rebooted this laptop, and saw that indeed, a kernel update was being offered, which I immediately installed. So that laptop now has kernel version ‘4.4.0-30-generic‘, or so my /boot directory would seem to say.

One problem that I was experiencing with that laptop since before camping, was some subtle WiFi issue which I could no longer pinpoint. I had written, that its ability to use the hardware encryption offered by the (kernel module ‘RTL8723BE’) chip-set seemed to work fine. But there were some other problems with the WiFi.

I would like to be able to report, when and if that issue has been resolved completely. But since Klystron has only been running on kernel version 4.4.0-30-generic for one day, it is still far too soon to call out a victory. I will continue to observe the behavior of that laptop for the next little while, and give further comment on it later. So far its behavior looks good.



Progress with the WiFi on Klystron

In This Earlier Posting, I had written that there were once stability issues, with the WiFi chip-set of my laptop, which I name ‘Klystron’.

Since then, there has been some updating of the kernel, and therefore also some updating of the in-tree kernel module, which operates the WiFi chip-set, and some updating of the firmware as well.

The way the behavior of the laptop has been since then is, that its WiFi connection finally seems stable, at 802.11n speeds and with H/W encryption, but that a rather odd behavior has cropped up, according to which the membership of that laptop in my Samba Workgroup, seems to fail overnight, consistently.

What I seem to decipher, is that this later malfunction is related to the fact that, like many Linux systems, ‘Klystron’ restarts some of its services periodically – nightly – including the ‘nmbd‘ Daemon, the Linux equivalent of the ‘Network Monitoring BIOS Daemon’, which emulates ‘NetBIOS‘. And something about the way the restart takes place is still not perfectly kosher. It seems to require I disconnect from the WiFi afterward, and simply reconnect, before the local laptop shows up as part of my Workgroup again. And yet, it remains fully possible to surf the Web, as well as just to connect directly to my Samba server and to browse it, using a predefined IP address…

But for the moment, there is no reason to suspect, that this has anything at all to do with the Firmware or the Kernel Module. And it has taken me some period of observation to reach that conclusion.