Re: strange file query
On 17/01/2022 10:07, ghe2001 wrote: Looking at your link to Wikipedia, it does indeed seem to have something to do with Mesa (there's a paragraph in there about 'shader'). But I've never heard of Mesa, and I certainly didn't install it. This sounds like one vote for deletion. Or maybe .8 votes :-) No. Leave well enough alone. Mesa is a core part of the graphics stack. Heaps of Debian applications use OpenGL for hardware-accelerated graphics rendering. Shaders are tiny programs that run on your GPU. If you delete this cache, it will likely be recreated, or Bad Things May Happen. You will have many Mesa libraries installed as dependencies: dpkg -l "*mesa*" On my system (sid), this removes 100 packages (note "-s" to simulate), including xserver-xorg!: apt-get purge -s -V libglapi-mesa There are likely other direct dependencies. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: po...@lists.debian.org
On 10/01/2021 11:33, gru...@mailfence.com wrote: and can i do that in stretch or do i need buster Thanks you so much. This is by far the best contribution to this thread. :-D Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: xfce widget problem
On 11/11/2020 11:26, ghe2001 wrote: How can I remove a widget that doesn't have a right-button xfce menu (the one with 'Remove' in it)? There's lots of info on the web about sticking one in the panel, but I can't find anything about removing one. Right-click / Panel / Panel Preferences / Items to see what you have in that panel. If it has no Remove option, it likely is a child of Notification Area that has been added by a process. I do not recommend removing Notification Area. Most applications that do this have an option to disable showing an icon in the "System Tray" (Xfce Notification Area is an implementation of this concept). As a last resort, you can exit the application or stop it starting in Sessions / Settings and Startup. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: title bar can't show chinese
On 17/09/2020 10:16, Long Wind wrote: firefox can show chinese page except in title bar(though in each tab's window, title is correctly shown) i use twm and below is option in /etc/X11/twm/system.twmrc TitleFont "-adobe-helvetica-bold-r-normal--*-120-*-*-*-*-*-*" i delete line above, it doesn't help both stretch and buster have this problem twm is ancient, so you have to specify an X11 bitmap font with Chinese language support. Consider using a window manager with scalable font support (this is what Firefox uses to draw its content). Xfce is a popular and lightweight choice. But twm can be made to work: I copied the system twmrc to ~/.twmrc and edited the TitleFont line to: TitleFont "hanzigb16fs" and then restarted twm. The Firefox window title was then displayed in Chinese for a Chinese language page. In fact, all window titles were displayed in Chinese. Tested with debian-live-10.3.0-amd64-xfce.iso under QEMU with locale zh_CN.UTF-8 set with "dpgk-reconfigure locales" and twm manually installed. Here are a few Chinese bitmap fonts installed by default: $ xlsfonts | grep gb2312 -isas-fangsong ti-medium-r-normal--0-0-72-72-c-0-gb2312.1980-0 -isas-fangsong ti-medium-r-normal--16-160-72-72-c-160-gb2312.1980-0 -isas-song ti-medium-r-normal--0-0-72-72-c-0-gb2312.1980-0 -isas-song ti-medium-r-normal--16-160-72-72-c-160-gb2312.1980-0 -isas-song ti-medium-r-normal--24-240-72-72-c-240-gb2312.1980-0 These three are aliases for the three fonts above without --0-0 in their names $ xlsfonts | grep hanzigb hanzigb16fs hanzigb16st hanzigb24st Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Create 3D text?
On 10/09/2020 13:53, Carl Fink wrote: Can anyone suggest a Debian repo-installable program to create 3D text? Blender. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: passwords + bad memory - Was (Re: how to test disk for bad sector)
On 02/09/2020 06:42, Mike McClain wrote: On Mon, Aug 31, 2020 at 09:41:06PM +, Long Wind wrote: my memory is poor, i can't remember many accounts and passwords The more experience you have the harder it is to find the memory you're searching for. That's my story and I'm sticking to it. Yes, you can't pour a gallon of knowledge into a shot glass without spilling a few drops (for those of us old enough to remember Married with Children). I use and recommend KeePassXC under Debian and KeePassDroid under Android to manage passwords without online services. They use the same .kdbx file format. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: recently-used.xbel
On 21/08/2020 12:31, Ben Caradoc-Davies wrote: On 21/08/2020 14:01, Mike McClain wrote: I took a look at ~/.local/share/.recently-used.xbel and see that not only is it tracking what I do but claims to be the property of freedesktop.org. This file records your most recently used local files. It is used to populate the "Recent Files" lists in many applications. If this still bothers you, you can disable it, but this seems to require per-toolkit settings. For example, for GTK3, edit ~/.config/gtk-3.0/settings.ini and add this line to the [Settings] section: gtk-recent-files-enable = 0 See: https://unix.stackexchange.com/questions/74031/disable-recently-used-in-gtk-file-directory-selector Delete ~/.local/share/recently-used.xbel then log out and back in. You might need similar settings for other toolkits such as GTK2 or Qt; the reappearance of ~/.local/share/recently-used.xbel will alert you to this. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: recently-used.xbel
On 21/08/2020 14:01, Mike McClain wrote: I took a look at ~/.local/share/.recently-used.xbel and see that not only is it tracking what I do but claims to be the property of freedesktop.org. Is there any way to see if this is being sent to them and who might they be selling this info to? It's bad enough that Google and so many others care to spy on us but it's really uncomfortable to think my own computer is doing it. This is worse than Big Brother. This file records your most recently used local files. It is used to populate the "Recent Files" lists in many applications. It is not sent to anyone unless you are syncing your home directory with a cloud service. Every href in my ~/.local/share/recently-used.xbel starts with "file:///" (a local absolute file URL). freedesktop.org is the organisation that maintains standards for filesystem layouts and some configuration file formats so that different applications and Linux distributions can use the same directory structure, launchers, and so on, to make life easier for users and application developers. They never receive user data. The tag owner="http://freedesktop.org;> in recently-used.xbel just means that they define the meaning of the enclosed tags. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Recommendation for filesystem for USB external drive for backups
On 20/08/2020 10:08, David Christensen wrote: On 2020-08-13 01:31, David Christensen wrote: Without knowing anything about your resources, needs, expectations, "consistent backup plan", etc., and given the choices ext2, ext3, or ext4 for an external USB drive presumably to store backup repositories, I would also pick ext4. If you want to access the backup drive from foreign operating systems: If interoperability is a consideration, FAT32 and NTFS should also be considered. FAT32 is widely used for removable flash media but has a 4GB file size limit, no journaling, and no support for permissions. NTFS is widely used for external hard drives and has journaling and support for attributes. If you buy a consumer-grade external hard drive, it will most likely be formatted with NTFS. Backup archives (such as tar archives) can be used to preserve Linux file metadata (permissions and timestamps) on foreign filesystems. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Recommendation for filesystem for USB external drive for backups
On 13/08/2020 12:14, rhkra...@gmail.com wrote: I'm getting closer to setting up a consistent backup plan, backing up to an external USB drive. I'm wondering about a reasonable filesystem to use, I think I want to stay in the ext2/3/4 family, and I'm wondering if there is any good reason to use anything beyond ext2? I use and recommend ext4. Journaling protects against filesystem metadata corruption, which can be caused by an electrical outage or system crash. ext3 also has journaling, but I see no reason to not use ext4. It is robust, widely deployed, and the default in Debian. My backups are pigz-compressed tar archives, encrypted with gpg symmetric encryption, with a "pigz -0" outer wrapper to add a 32-bit checksum wrapper for convenient verification with "gzip -tv" or similar without requiring decryption. Archives are written to both external local storage and uploaded to cloud storage. Because I have a small number of large backup files, I make backup filesystems to optimise for large files and maximise space: minimal journal, no reserved blocks, large file inode ratio, and no support for resizing while mounted. I also disable mount count and interval checking: mkfs.ext4 -J size=4 -m 0 -T largefile4 -O "^resize_inode" /dev/sdb1 tune2fs -c 0 -i 0 -L Backup /dev/sdb1 I have this line in /etc/fstab: LABEL=Backup /media/backup ext4 noatime,noauto,user,errors=remount-ro 0 0 Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: what calculator do you use?
On 14/07/2020 03:39, Wolf Halton wrote: I know you are looking for a GUI calculator app, however I would like to entice you with a CLI app that is easy to use, but will make your coworkers think you are secretly a computer expert. I tend to use python3 in the terminal window. I open a terminal, type python or python3 in the command line. It lets me do relatively complex calculations, and makes easy copy/paste of the entire calculation where required I use powershell in Windows the same way. Can just use the up-arrow to repeat similar calculations. Python is a programming language and it’s pretty simple to set up all kinds of repetitive math quite easily, but you probably won’t need all the capability (at least at first). 4/5 (enter) gets you 0 which is good for estimating but 4.0/5 gets you 0.8 like you probably expect "Floor division" is the default in Python 2 (without "from __future__ import division"), but in Python 3, "true division" is the default and there is a new "floor division" operator. Python 2: 4 / 5 == 0 4.0 / 5 == 0.8 Python 3: 4 / 5 == 0.8 4 // 5 == 0 Project Jupyter <https://en.wikipedia.org/wiki/Project_Jupyter> provides a rich set of tools and supports Python. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: what calculator do you use?
On 13/07/2020 19:31, kaye n wrote: Correct me if I'm wrong, but I don't see any calculator app in my Debian os. What do you guys use? I'm having trouble with Galculator. Python 3 with the contents of the math module imported. I have this in my ~/.bashrc: alias pyc="python3 -i -c \"from math import *\"" Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Thank you [was: Reminder about the Debian Code of Conduct]
On 26/06/2020 02:50, Default User wrote: "Beware the censor, for in his heart, he deems himself your master." main! -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: why foxit hasn't been included in buster?
On 25/06/2020 17:01, Long Wind wrote: oh, i see, it's not free, so it isn't included in debian mainthen i will try evince qpdfview is also nice. Qt application so takes a little fiddling to get consistent theming if you are using Gnome or other GTK-based desktop. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: technical terms overhaul
On 20/06/2020 11:19, Larry Martell wrote: On Fri, Jun 19, 2020 at 7:08 PM Eike Lantzsch wrote: I really like "redlist" and "greenlist". Offensive to Native Americans and Martians. And also culturally confusing to some Chinese in which red is a happy colour. As I understand it, Chinese stock market colour coding is opposite to Western. Red/green is also the most common form of colour-blindness. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: technical terms overhaul
On 20/06/2020 09:45, davidson wrote: On Fri, 19 Jun 2020, Dan Ritter wrote: Most things described as master/slave aren't actually very well matched by that metaphor. Source/sink, boss/worker, Woah there. "Team Leader" / "Team Member", please. "Secretary of the Workers' Revolutionary Committee" was once popular, but has not yet, as far as I know, been applied to technology. Some high-availablity clusters vote to elect their leaders so that a defective primary can be deposed. "Associate" is a popular weasel word for lowly-paid employees. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: technical terms overhaul
On 20/06/2020 07:14, Eike Lantzsch wrote: instead of (for example on disk drives) "master" and "slave" I like to propose the terms This discussion reminds me of the PC gaming supremacists on the pcmasterrace subreddit <https://old.reddit.com/r/pcmasterrace/>. The term "master" has a long history. In addition to its racial connotations, it is also gendered. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Cannot open Trash in spacefm?
On 07/03/2020 07:40, kaye n wrote: *What happens if, as a test, you select Thunar as your preferred file manager, and then double-click on the Trash icon?* It opens! This confirms that the problem is in spacefm, not Thunar or Xfce. I was able to reproduce the behaviour you report on a buster live image after installing spacefm. I am not familiar with spacefm, but seems that it does not support Trash, trash:/// urls, or any other part of the Freedesktop.org Trash specification. There are some community plugins for Trash support, but I think you also need handler support to get trash:/// urls to work. I have updated the title to draw attention from anyone who might be able to help with spacefm. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Cannot open Trash?
On 06/03/2020 16:28, kaye n wrote: I forgot to mention that I've made spacefm the default file manager of my system. Double-clicking the trash icon in the desktop screen and then choosing a file manager to open it results in /home/user being shown, not the trash directory, regardless if I choose thunar or spacefm. What happens if, as a test, you select Thunar as your preferred file manager, and then double-click on the Trash icon? I have not tried spacefm. What happens if you try to open Trash directly with spacefm? spacefm trash:/// It is possible that exo-open is misconfigured (and I think that is what is used to open preferred applications). For me, this opens Trash in Thunar (my preferred file manager): exo-open --launch FileManager trash:/// Please also see the following output: *kaye@laptop:~$ sudo dpkg -l "gvfs*" It looks like you have all the expected gvfs* packages so this is not the problem. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Cannot open Trash?
On 06/03/2020 11:39, Ben Caradoc-Davies wrote: I tried purging gvfs-backends and was still able to access Trash via Thunar, so missing this package is not likely to be your issue. But if I purge gvfs itself, Trash vanishes from Thunar and cannot be accessed by manually entering the location "trash:///". If you are missing the gvfs package, you can install it with: apt-get install gvfs or via the package manager of your choice. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Cannot open Trash?
On 06/03/2020 11:29, Ben Caradoc-Davies wrote: What gvfs packages do you have installed? See with: dpkg -l "gvfs*" Some gvfs-* packages are optional and one might include the handler for trash: urls. apt-get dist-upgrade on sid recently tried to remove gvfs-backends during the python3.8 transition, but you are on buster so should not have been affected, and I do not think gvfs-backends provides the handler for trash: urls. I tried purging gvfs-backends and was still able to access Trash via Thunar, so missing this package is not likely to be your issue. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Cannot open Trash?
On 05/03/2020 20:55, kaye n wrote: Hello Friends! My system: Host: laptop Kernel: 4.19.0-6-amd64 x86_64 bits: 64 Desktop: Xfce 4.12.4 Distro: Debian GNU/Linux 10 (buster) There is a Trash icon on the desktop screen. If I double-click it to open it so I can view its contents, I get a dialogue box with the title, *Handler Not Found* The message of the dialogue box is: *No network handler is configured for this URL, or no mount command is set. Add a handler in Devices|Settings|Protocol Handlers.* I'm too ignorant right now to figure it out. I would appreciate it if anyone could give a step by step instruction. Thank you very much! Can you open Trash in Thunar? If not, try opening a Terminal, quit Thunar with Close All Windows (or use "thunar -q" as I am not sure quit works if Thunar is managing your desktop), run "thunar" at the command line, and look for error messages. You might also see error messages in ~/.xsession-errors . What gvfs packages do you have installed? See with: dpkg -l "gvfs*" Some gvfs-* packages are optional and one might include the handler for trash: urls. apt-get dist-upgrade on sid recently tried to remove gvfs-backends during the python3.8 transition, but you are on buster so should not have been affected, and I do not think gvfs-backends provides the handler for trash: urls. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: The nightmare of Intel Integrated GPUs under Linux in general and Debian in particular
On 13/02/2020 18:24, T wrote: Have you monitored your CPU temperatures? mprime (AKA Prime95) is a well-known CPU stress tester. I would also test memory with multiple concurrent instances of memtester. All problems I suspected of being caused by CPU overheating were in fact caused by RAM (most recently, a tiny overclock that seemed safe and harmless but caused intermittent corruption). Integrated GPUs are totally reliant on system RAM. There is also gputest, but I have not used it. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Best file system to use?
On 13/02/2020 12:55, Dennis Wicks wrote: I have 4TB running on an AMD Ryzen under Buster. What is the current consensus of the best file system to use for general data usage? I have been using xfs but that is based on info from many years ago. With apologies to Samuel L Jackson (as Beaumont in Jackie Brown): ext4. The very best there is. When you absolutely, positively got to have the most robust and widely deployed Linux filesystem, accept no substitute. In all seriousness, unless you have niche performance needs and quantitative evidence to support them, or require fat32 for interoperability (and are willing to sacrifice Unix permissions and file size), do not overlook the robustness and ubiquity of ext4. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: buster: low audio level
On 13/02/2020 13:48, Jonas Smedegaard wrote: Quoting Doug McGarrett (2020-02-13 01:15:57) What on earth is bikeshedding? That's a new one on me! That's when you ask something in a large community that is easy to have an opinion on and with many possible opinions - e.g. asking "which color should we paint our bikeshed?" or "what disk format is best" or "which computer should I buy?" https://en.wikipedia.org/wiki/Law_of_triviality The choice of easy problems with many possible solutions is the observed behaviour, but the outcome is that "members of an organization give disproportionate weight to trivial issues". Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: apple mini
On 09/01/2020 16:45, David Wright wrote: No, don't mix degaussers and disks. If you want to reuse them, they're likely too damaged. If you're concerned about data recovery, then they're unlikely to be erased enough to prevent it. <https://en.wikipedia.org/wiki/Degaussing>: "For certain forms of computer data storage, however, such as modern hard disk drives and some tape drives, degaussing renders the magnetic media completely unusable and damages the storage system. This is due to the devices having an infinitely variable read/write head positioning mechanism which relies on special servo control data (e.g. Gray Code) that is meant to be permanently recorded onto the magnetic media. This servo data is written onto the media a single time at the factory using special-purpose servo writing hardware." -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: apple mini
On 09/01/2020 15:57, ghe wrote: Back in the analog days, I worked at a college radio station that sent out radio programs on tape. There was a big box that we passed a reel of tape over to erase it. That box might do disks too :-) I recall reading that the information encoded on HDDs is much more densely and strongly encoded and on multiple platters, so much harder to reliably erase with external magnetic fields such as those generated by a bulk tape eraser. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: apple mini
I use ATA secure erase <https://ata.wiki.kernel.org/index.php/ATA_Secure_Erase>; I activate it with hdparm from a live usb drive (you may need to install hdparm). I do not know if ATA secure erase is supported by the mini bios and drive. Many bioses I have seen that freeze security can be unfrozen with a suspend/resume cycle. ATA secure erase is fast, optimised for the drive, and can also erase unallocated blocks that are not visible to the operating system (especially those on SSDs). No HDD can be fully overwritten in two minutes. Enhanced erase may offer more protection, but you are at the mercy of the HDD vendor implementation. I also encrypt all my drives. If you need to protect against an attacker willing to examine your HDD with magnetic force microscopy, there is no substitute for physical destruction of the media. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Firefox equivalent of SeaMonkey's pref.js
On 20/12/2019 10:52, Richard Owlett wrote: As I date from Netscape era I much prefer SeaMonkey. However when I did a reasonably default install of Debian 9.8 I got Firefox 60.5.0esr . Due to *OPERATOR ERROR*, I mangled a preference. On the mozilla.general list I was told to edit pref.js . For SeaMonkey I have no problem. For Firefox where/what is equivalent file/directory? ?? ??? ?? TIA prefs.js (with an "s"): ~/.mozilla/firefox/$PROFILE/prefs.js where $PROFILE is the generated profile directory. If you have multiple profiles, you can find all instances of prefs.js with: find ~/.mozilla/firefox -name prefs.js Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: USB Examiner Package? Special USB Kernel Modules?
On 27/11/2019 04:28, Kenneth Parker wrote: My suspicion, by the way, is that the Power types of Malware are caused by shorting out some Pins. USB is designed to survive shorts. An example of a malicious device is the USB Killer, which is available from a commercial supplier. Similar devices have been improvised for very little cost: https://en.wikipedia.org/wiki/USB_Killer "The device collects power from the USB power source of the component it is connected to in its capacitors until it reaches a high voltage and then it discharges the high voltage onto the data pins." [...] "In April 2019, a 27-year-old Indian former student of the College of Saint Rose, Vishwanath Akuthota, pleaded guilty to destroying 59 computers in his college using a USB killer, resulting in over $50,000 in damages." Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: USB Examiner Package? Special USB Kernel Modules?
On 26/11/2019 16:37, Kenneth Parker wrote: Here's an interesting one: A Windows friend handed me a USB Dongle, knowing that I'm a Linux user. He says he got it 3rd hand, with info that it might be "Very Dangerous". He would be interested, if I find out something about it. (And, indeed, Google has many hits on "USB Malware"). So, what I want, is a USB Debugging Package, that will *NOT* attempt to, actually open this device, but will give me information about it. Obviously, this has to be handled carefully because, for one, it's not always obvious which USB goes where. For example, before I plug it in, "lsusb" should not show anything plugged in. End of preliminaries. When I plug in something, (i.e. Serial Mouse in Text Only environment, or a USB Thumb Drive), a Flurry of Activity ensues, with lots of Kernel Messages (and before I get to examine it). Does that mean I have to make a Custom Kernel for this, or limit the Kernel Modules used? Any insights so far? I take it that you are aware that there are malicious USB devices that look like thumb drives and can: (1) impersonate an HID (e.g. keyboard), fingerprint the host, open a terminal, and start typing at maximum rate to download malicious software (hence your interest in disabling kernel USB support), or (2) deliver a high voltage to the USB bus to inflict physical damage. At the least, I hope you have watched all USB-related DEFCON videos on YouTube, especially those on BadUSB. One or two were enough for me to never want to use a USB thumb drive of unknown provenance. That is all I know. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Orphaned Inode Problem
On 19/11/2019 08:17, Reco wrote: A kernel panic or OOPS comes to mind first. That's very broad class of the problem, to say the least, hence the need of kernel logs. Xorg hang is the second possible option. AMD hardware is somewhat problematic here. Barring above - an overheat is the third possible scenario here. Memory errors under load can also cause kernel panics. I like to test RAM by running multiple concurrent memtester processes to simulate multithreaded load. Some AMD platforms had a reputation for instability with anything less than the highest performance RAM. mprime (Prime95) is useful for stress testing a CPU. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Top 7 Programming Languages That Employers Really Want
On 18/10/2019 15:33, Turritopsis Dohrnii Teo En Ming wrote: This is just a quick survey. May I know what programming languages do you know? I am considering being a programmer or developer. How long will it take for me to master a programming language like C++, Java, and Python? Mastery is a high level goal. Are you trying to learn enough to secure employment in an entry-level position? Python is easiest to learn and widely used, but not commonly used in packaged products. Java is the most widely used, and designed as a more portable and easier to learn alternative to C++, which is also widely used. C++ is by far the hardest. Do you have a particular industry in mind? The choice of language will follow the domain. Software development is about much more than just programming: teamwork, engineering practices, and personal technical and business problem-solving ability. What is your background? A great way to learn is to choose an open source project that uses a language of interest to you and contribute bug fixes to it. Proven ability on open source projects is a great way of building your resume. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Screensaver issues after distro upgrade
On 11/09/2019 06:11, Christopher David Howie wrote: On 9/10/19 4:30 AM, Miroslav Skoric wrote: After upgrading the old laptop from Jessie to Stretch, I noticed that the screensaver in Mate environment does not work for me as before. For example, when the screen goes black after some time of inactivity, for returning back it is not enough just to touch the touchpad or press any key. Instead, pressing any key or touching the touchpad makes the screen just some 1% lighter than the full black (or better to say, it remains 99% black). However, the last working GUI does not return. I have this same issue with the default configuration of XFCE on Buster, where the problem was not present on Stretch. Is light-locker installed, and do you use lightdm? light-locker appears to be the source of this issue. Googling "light-locker black screen" returns dozens of posts across many sites complaining about the same problem. I resolved this issue by removing light-locker and installing xscreensaver instead. Note that this required removing a few metapackages, and then marking the dependent packages that I wanted to keep as manually installed to prevent apt from removing most of my desktop tools. I saw the same problem when I tried lightlocker after upgrading to XFCE 4.14, so I went back to good old xscreensaver. Ugly bitmaps fonts and no theming, but secure and reliable. There is a new locker in XFCE, but there have been reports of segfaults causing uncommanded unlock, so I consider it too vulnerable for use. There is a bug report. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Xfce 4.12 to 4.14: high Xorg CPU usage
On 28/08/2019 01:01, Stefan Pietsch wrote: I upgraded Xfce to 4.14 recently (Debian unstable) and noticed slightly delayed rendering of UI elements. Firefox and Thunderbird behave sluggishly and CPU usage by Xorg is significantly higher as compared to Xfce 4.12. Did anyone who is using Xfce 4.14 observe similar effects? Does top/htop show any process burning CPU? I mask the colord service because I do not need it, and after upgrading from Xfce 4.12 to Xfce 4.14, my session acquired an xiccd process that sat burning 100% of one CPU (unhappy about not being able to contact colord?). I disabled it from my Xfce session in Settings / Session and Startup / Application Autostart, logged in and out, and the problem went away. I do not save my sessions. Other than needing to use ~/.config/gtk-3.0/settings.ini to force all Xfce GTK 3 elements to use the correct font size (the setting is ignored for some elements), everything else works for me as well as in 4.12 and performance is satisfactory. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: duckduckgo
On 19/08/2019 08:52, to...@tuxteam.de wrote: This is actually for me a filter criterion: if a site doesn't work with javascript, chances are high that I avoid it. I do make some exceptions, but very few. I use NoScript to enable JavaScript only where I want it. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: How free is Debian
On 09/08/2019 12:05, John Hasler wrote: There is a lot wrong with the patent system. Twenty years is too long. Fees are too high. Processing is too slow. The language used in the disclosures is arcane (the disclosure is supposed to teach the invention to someone "skilled in the art", but the art one must be skilled in is patent law). Software patents not only exist, but are described in *incredibly* arcane language rather than in source code. Pharmaceutical patent law is even more bizarre than software patent law (mostly due to the efforts of Congress to "fix" it). New Zealand banned pure software patents in 2013: How New Zealand banned software patents without violating international law https://qz.com/119419/how-new-zealand-banned-software-patents-without-violating-international-law/ Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Wireless home LAN - WiFi vs Bluetooth?
On 01/08/2019 01:44, Reco wrote: On Wed, Jul 31, 2019 at 02:32:25PM +0100, Brian wrote: On Wed 31 Jul 2019 at 16:07:33 +0300, Reco wrote: On Wed, Jul 31, 2019 at 07:58:54AM -0400, Celejar wrote: mathematical analysis of how much hardware would be necessary to crack a good WPA2 password. I've seen lots of sites explaining how to use hashcat with a GPU, and various real-world tests on lists of hashed passwords (e.g., [1]), but can you provide a serious analysis of the practical cost, in time or hardware, of cracking a real-world WPA setup? Cost - Amazon will take 11c per hour for that VM that comes with NVIDIA Tesla videocard. Said hour is more than enough to bruteforce 8 character WPA passphrase with hashcat. In the context of a home user producing a secure wireless configuration, a 64 random character passphrase works wonders. The sky is not about to fall in. Agreed. If 64 character password is reasonably random, bruteforcing it is economically unfeasible. With obvious exceptions, of course. Entering such password to a device is somewhat tedious though. Especially tedious on devices with limited input interfaces, such as smart TVs, game consoles, and printers. Restricting the WPA2 passphrase to digits and lowercase letters reduces entropy but makes input more bearable. Seeing the reaction of guests when they are handed a piece of paper with a long random WPA2 passphrase: priceless. It never gets old. :-D Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Can't install addons for firefox
On 05/05/2019 10:52, Ben Caradoc-Davies wrote: Setting about:config / xpinstall.signatures.required to false fixed extensions for me for firefox 66.0.1-1 amd64 on Debian sid and for Firefox 66.0.2 on Android. I needed to restart Firefox on Debian. Seems fixed in firefox 66.0.4-1 on Debian sid and firefox 66.0.4 on Android: extensions continue to work after restoring the default about:config xpinstall.signatures.required setting (true). Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Can't install addons for firefox
On 05/05/2019 21:30, Dave Sherohman wrote: On Sun, May 05, 2019 at 10:52:56AM +1200, Ben Caradoc-Davies wrote: Setting about:config / xpinstall.signatures.required to false fixed extensions for me for firefox 66.0.1-1 amd64 on Debian sid and for Firefox 66.0.2 on Android. I needed to restart Firefox on Debian. When I tried that yesterday, it allowed extensions to be reinstalled (yes, I was one of those who tried to fix this problem by uninstalling and reinstalling my extensions), but not all plugins actually worked in this state. Most notably, NoScript displayed no icon on the toolbar (just a blank space with a tooltip) and did not appear to actually do anything. Its settings were also inaccessible, both through the (missing) popup menu and via the list of installed add-ons (in which case I got a blank grey page with an endless "loading" spinner animation). Based on this, it appears that xpinstall.signatures.required may be no more than a partial solution. (Although it's also possible that something may have been broken in the uninstall/reinstall process.) My prexisting NoScript installation works correctly after setting xpinstall.signatures.required to false, including Add On preferences page and toolbar icon. I did not uninstall/reinstall. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Can't install addons for firefox
Setting about:config / xpinstall.signatures.required to false fixed extensions for me for firefox 66.0.1-1 amd64 on Debian sid and for Firefox 66.0.2 on Android. I needed to restart Firefox on Debian. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Using USB Flash / Pendrives or [micro]SD cards for backup
On 19/03/2019 02:30, rhkra...@gmail.com wrote:
Is anyone (reading this list) using USB Flash / Pendrives or [micro]SD cards
for backup?
I've thought about doing that, especially as they continue to come down in
price, but my experience with them at least in some cases has not been good.
The worst case seems to be dash cams where, after about a year or so they just
stop working, but I've had at least one similar case using them for data
storage.
I have six 16 GB ext4-formatted USB 2 flash drives of two models that I
rotate for daily local backups. Each backup is about a 3 GB
gzipped-gpged-gzipped-tar file, so the write pattern is large sequential
writes. The outermost zero-compression gzip ("pigz -0") allows a keyless
32-bit checksum test with any gzip implementation. Each thumb drive
contains no more than three backup files at a time. All are
checksum-tested after each backup.
I have been using these drives for at least five, and perhaps more than
seven years. My rough estimate is that I have written around 60 times
their capacity to these drives. I have had one drive fail, about five
years ago.
My rules for USB flash backup are: (1) nothing gets written to USB
without first being encrypted because failed drives cannot be erased
without physical destruction, and (2) nothing gets written to USB
without a checksum because USB drives are unreliable and will happily
and silently corrupt data.
Redundancy and the expectation of drive failure will reduce
inconvenience when it occurs. And it will. But you have to be able to
detect it.
I also have offsite backup in the cloud with the same encryption and
integrity measures (plus cloud encryption), and retention of longer-term
archival backups. Now that I have gigabit fibre, offsite backup is
faster than local.
Kind regards,
--
Ben Caradoc-Davies
Director
Transient Software Limited <https://transient.nz/>
New Zealand
Re: ssh policy change in new stretch??
On 08/03/2019 12:12, Long Wind wrote: i run command below and it work in early debian: ssh zhou@192.168.1.3 but new stretch says: ssh: connect to host 192.168.1.3 port 22: Connection refused what should i do? Thanks! Has the server also been upgraded to stretch? Is sshd running on the server? Is there a firewall on the server? Try: ssh -vvv zhou@192.168.1.3 to see the full connection attempt. You can also check whether the port is open with (on the client): nmap -p 22 192.168.1.3 Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Why popular sites are looking ugly
On 24/01/2019 01:07, Marek Mosiewicz wrote: After some experiments with dpkg-reconfigure fontconfig-config it seems that the issue is with native/autohinting option. Seems that Debian uses Dejavu fonts which looks better with native. I use manually hinted fonts, including Liberation Sans and DejaVu Sans Mono, with full greyscale hinting at 96 dpi. I do not like the rendering of these fonts with the automatic hinting of the new v40 TrueType interpreter, which is the default in FreeType 2.7 or later: https://www.freetype.org/freetype2/docs/subpixel-hinting.html I enable the old v35 interpreter everywhere by default with the following in /etc/environment: FREETYPE_PROPERTIES=truetype:interpreter-version=35 The new interpreter is supposed to have better results with MS fonts, so this may not fix your problem. For lovingly hand-hinted fonts on low DPI screens (mine is physically about 82 dpi (27" 1080p)), the old interpreter might be worth a try. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Log in details
On 12/01/2019 16:05, Johndy Laviña wrote: Hi, May I know what is the login and password for Debian GNU / Linux 9? Thanks Live or installed? For Debian Live, the login is "user" and password is "live". You will need these if you lock the screen. To get root, open a terminal and use use "sudo -i". For installed Debian, login and password are set at install time. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Looking for advice on tools (or libraries) for unsupervised, bulk symmetric encryption/decryption of files
On 10/01/2019 03:05, Kynn Jones wrote: The only encryption tool I have used for encrypting files on my hard drive is gpg2, which I have used for small, interactive encryption tasks (half-dozen files, at most). Therefore, my initial attempt was to use gpg2 for this new bulk-encryption task, but I found myself constantly fighting with it, and finally had to recognize that I was trying to use gpg2 for something it is not primarily designed for. (I am also a bit concerned with gpg2's future stability. AFAICT, It's design has varied significantly over the years, and as a result there's a lot of confusion on its use. That has been my experience, in any case.) I use a pipe with gpg2 as one component for symmetric encryption: gpg --batch --symmetric --cipher-algo AES256 --s2k-digest-algo SHA512 --compress-algo none --passphrase-file $PASSPHRASE_FILE My pipe input is usually a tar file gzipped with pigz for parallel compression, hence the "--compress-algo none". I then add another "pigz -0" wrapper to get a cryptographically weak checksum to allow testing for media failures without the passphrase. I like tar because it preserves file metadata and filesystem structure and is a very stable format. Other formats may be better for random access. Recently I used gpg2 to decrypt files that were encrypted over 15 years ago; note that these were much smaller files and a simpler invocation of gpg1 (the then default cipher was CAST5 IIRC). The gpg file format seems well-documented and stable. Regular decryption tests are prudent to catch problems after gpg upgrade. Yes, the new interactive predilections of gpg2 were a pain at first when compared to gpg1, but "--batch" and "--passphrase-file" seem sufficient for batch symmetric encryption, if you do not mind your passphrase being in plain text on your filesystem. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: do you find old firefox is better than new one?
On 17/12/2018 06:35, dekkz...@gmail.com wrote: On 12/16, Long Wind wrote: i have 52.9.0 and 45.9.0, both for stretch new one often becomes unresponsive, and i have to close it and restart it New FF is for me by far the better, quicker and far better memory use. Firefox 63 and 64 have been good for me. Earlier 60s seemed to have slow startup but this might have been because of plugins that have now been improved. I have not experienced any hangs or resource problems, and I use many windows and tabs. I use NoScript and Adblock Plus, which reduce resource consumption. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: question about usb socket on front panel
On 12/12/2018 14:04, Long Wind wrote: sorry! this question isn't linux specific there are two usb sockets on front panel of pc casethey appear same, but really are notwireless mouse can only be connected to one of them could you tell me their difference? Thanks! Bad connection on one port? USB front ports are often connected to the motherboard by a pair of internal cables. A bad connection would prevent the mouse from working. This could be caused by a flipped internal connector, loose connection, or bad cable. A flipped connection can prevent booting by shorting 5V to ground. Do other USB devices work on the port that does not work with the mouse adapter? Sometimes USB ports differ by speed (blue for USB 3, black for USB up to 2), or permit USB charging (lightning bolt symbol, supports higher current). Neither should affect a wireless mouse, because it is a low speed device, unless the mouse adapter just will not work with a hub at using a particular specification. The hub is internal and typically on the motherboard. The front USB ports I have seen are connected to the same hub so I would expect both to work or neither. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Reverting firefox-esr upgrade in Buster
On 09/11/2018 03:53, Cindy-Sue Causey wrote: But... bank account STILL does not work... It yelled javascript was not enabled and basically*almost* KICKED me out IMMEDIATELY because it wasn't available. Which led to me sitting here thinking... I THOUGHT javascript was security issue prone to the point folks are trying to wipe it from the Net. So WHY is a bank so dependent on it that my bank's website almost seemed to freak that javascript wasn't available??? I use and recommend Firefox with NoScript. Recent NoScript supports WebExtensions, with a new user interface. NoScript blocks JavaScript by default and protects you from a range of threats even for origins from which JavaScript is permitted. With NoScript, you can selectively enable JavaScript for those sites that you trust. I also use Adblock Plus because ads are a known delivery mechanism for malicious content. The combination of NoScript and Adblock Plus has the side benefit of making we browsing much faster. If you are paranoid, you can use a dedicated Firefox profile for your internet banking. If you are super-paranoid, you can run Firefox on a dedicated virtual machine. A freshly booted non-writeable live image for each banking session will prevent malicious state from being persisted on the client VM. As long as the host machine has not been compromised, in which case all bets are off. I like web sites that fail gracefully when JavaScript is disabled. Unfortunately, JavaScript Single Page Applications (SPA) are a plague on the internet. I prefer RESTful architectures. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Removing firefox-esr also removes gnome
On 06/11/2018 08:37, Reco wrote: Seems like other browsers like Iceweasel are just arbitrarily left out. It's been two years since they discontinued Iceweasel. Iceweasel was just Debian's Firefox rebranded to address trademark and logo licensing issues <https://en.wikipedia.org/wiki/Mozilla_software_rebranded_by_Debian#Iceweasel>. In June 2016, the iceweasel package was replaced with the firefox-esr package. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Help me Linux
On 30/10/2018 21:17, P M wrote: Although right now I am using Windows but still I feel very enthusiastic and energetic with Linux; even I don't know what the reason is. You are feeling the potential of open source: a community open to all and your freedom to change anything. Even Microsoft is embracing it. I TRIED TO INSTALL DEBIAN MANY TIMES BUT FAILED BADLY. I NEVER USED DEBIAN. Debian is harder to install than Ubuntu because it gives you more choice. Choice reduces happiness: https://www.ted.com/talks/dan_gilbert_asks_why_are_we_happy But choice gives you power. Try installing a minimal Debian in a virtual machine. Try adding more packages and a desktop environment. suggest me what can I do? how should I proceed? how should I go further in the field of Linux? and what can I do with and how to go further in python? I recommend learning Linux and Python by using them wherever possible. What are your areas of interest? Games? Science? You could start a tiny Python project. A great way to learn from others is to contribute to an open source project. My one more query is that there is no any alternative available in linux for Corel draw, MS office, video editing software. Although there are many claims of alternatives but the reality is harsh, they are not as simple and productive as the windows software. Please clear my view regarding this also. I meet all my office needs with LibreOffice. If you prefer Windows offerings, you can run Debian in a virtual machine and have the best of both worlds. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: www-data
On 29/10/2018 11:42, mick crane wrote: I'm asking because somebody is saying that webmail server files should be owned by root but I don't know about that, if somebody as got so far to be www-data they might as well be root ? Web server configuration files are typically owned by root and not writeable by other users. Files containing secrets such a private keys may only be readable by root. When a server is started as root, it reads its configuration files and secrets, then drops privileges by changing user. During normal operation as an unprivileged user, the server cannot edit its own configuration files or write where access has not been granted to www-data. This provides a substantial level of protection that is absent in a server running as root. Apache can also be used as a proxy for other services such as Tomcat, providing an additional layer of protection. The specific case of webmail likely requires read-write access to user mailboxes. I do not know how privilege separation is handled in this case. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: www-data
On 29/10/2018 10:26, Carl Fink wrote: On 10/28/2018 05:16 PM, mick crane wrote: what's the deal with www-data ? I never made that user I dunno if it has a password or what ? these are things that some setup / install makes ? It's created by the Apache installer. Check the Apache docs. And I should mention that the www-data user can also be used by other web servers such as nginx, although apache is the usual suspect. The idea of having a common user for all web servers is to allow switching between server implementations without having to chown all your static content. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: www-data
On 29/10/2018 10:26, Carl Fink wrote: On 10/28/2018 05:16 PM, mick crane wrote: what's the deal with www-data ? I never made that user I dunno if it has a password or what ? these are things that some setup / install makes ? It's created by the Apache installer. Check the Apache docs. And it should have no password. This user is accessed by switching to it from root. As a security measure, after binding to privileged network ports as root, apache switches to user www-data so that, if it is compromised, the damage is limited. Processes that have dropped root privileges cannot automatically regain them. Postgres and Tomcat do the same thing with their own dedicated users. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: No Java in Eclipse Preferences
On 19/10/2018 03:48, Kent West wrote: I do feel a little tainted, going outside of the Debian repositories for software. But I understand that in the world of Free Software, there's not always a volunteer available to maintain a package. Fear not. The Eclipse Foundation has fine Open Source credentials. I use Debian OpenJDK packages but applications like Eclipse and Maven I install from tarballs and run with wrapper scripts so I can choose JDK versions. In the Java ecosystem it seems much more common to bundle dependencies with applications. This is quite different to the Debian way of doing things, and untangling these dependencies seems to have been an insurmountable obstacle to Debian packaging. I am not saying that it cannot be done, just that the cost is sufficient to prevent it with the amount of available developer interest. I have found Debian to be a good platform for Java development. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Unattended Upgrades question
On 16/10/2018 09:45, rv riveravaldez wrote: Hi, two doubts: 1. Is it OK to disable/remove Unattended Upgrades on Debian testing? 2. If the answer is 'yes': which would be the best/proper way to do it? Thanks a lot! I use systemctl to mask all the apt-daily units: systemctl mask apt-daily-upgrade.timer systemctl mask apt-daily.timer systemctl mask apt-daily.service systemctl mask apt-daily-upgrade.service I have not tried, but if you still want the apt update (download indices) but not the upgrade (install new packages), you might only need to mask the *upgrade* units. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Slow firefox and high cpu usage
On 11/10/2018 11:36, bw wrote: On Thu, 11 Oct 2018, Ben Caradoc-Davies wrote: On 11/10/2018 11:15, bw wrote: How exactly do you think stretch users should run an adblocker when all the xul-ext-* extensions are now broken? I see that there is a webext-ublock-origin for sid but I have never used it: https://packages.debian.org/sid/web/webext-ublock-origin p.s. and I use stable, because it is stable, not sid, which is unstable. thanks anyway but I think your advice is a little dubious. My point is not that you should use unstable, but that the evidence on sid suggests that webext-* packages are coming to stable ... when stable is called buster. I did not see any webext-* packages in stretch-backports. The workaround is to install them directly from upstream via Firefox. I agree that it is sad that Firefox on stretch has been upgraded to break the xul-ext-* packages before webext-* packages are available. Unfortunately Debian is wedged between upstream dropping support for xul-ext-* extensions in ESR 60 and the end of life of ESR 52. You do want security patches, don't you? I think that ESR 60 with unpackaged extensions is the lesser evil. Normal service will likely be resumed in buster. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Slow firefox and high cpu usage
On 11/10/2018 11:15, bw wrote: How exactly do you think stretch users should run an adblocker when all the xul-ext-* extensions are now broken? Install an extension built for webextensions such as Adblock Plus 3.0 or later using Firefox Add-ons Manager?: https://addons.mozilla.org/en-US/firefox/addon/adblock-plus/ I see that there is a webext-ublock-origin for sid but I have never used it: https://packages.debian.org/sid/web/webext-ublock-origin Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Slow firefox and high cpu usage
On 09/10/2018 07:07, Pétùr wrote: I experience a very slow response time (and high cpu usage) for firefox in debian sid these days. I use the 62.0.3 version and the latency is particularly present at the startup. I have to wait few seconds before being able to enter text in the address bar. I tried to reset (install new profile) firefox and launch it in safe mode (ie without extensions). No improvements. Are other users of sid experiencing the same behavior ? I have slow startup and a brief hang during initial UI layout, but only with Adblock Plus enabled. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: thunderbird - drag'n'drop now hides target?
On 08/10/2018 16:43, Richard Hector wrote: A nice trivial one ... after a recent update of Thunderbird, when I drag and drop a bunch of selected mails to a different folder, they now obscure the target folder - so unless I'm careful to drag by the very top edge, I can't see the target get highlighted. Obscured because the cursor changes to a list of the email titles? I disabled this by setting "nglayout.enable_drag_images" to "false" in Preferences / Advanced / Config Editor. This restored the previous behaviour in which only a simple drag cursor is displayed. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: question about ls
On 26/09/2018 10:55, Long Wind wrote: sorry! you're right. after checking ls manual, i find ls has option -1 Note also that ls will behave like it has the "-1" option if its output is piped to another command, even without this option. I like to use the "-1" option even in this case because it is what I mean (counting entries one per line). Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: question about ls
On 26/09/2018 10:27, Long Wind wrote:
can ls show number of items in a folder?Thanks!
I use:
ls -1 | wc -l
To include items whose names start with ".", add the "a" or "A" argument
to ls ("man ls" for details).
For more advanced searches (recursive or filtering on file type), I use
find.
Kind regards.
--
Ben Caradoc-Davies
Director
Transient Software Limited <https://transient.nz/>
New Zealand
Re: can apt-daily and apt-daily-upgrade be purged from Buster?
On 23/09/2018 13:12, Felix Miata wrote: What installs/owns these systemd pseudo-programs? I want them eradicated, not simply disabled. One or more of them by default lock package management at boot so that I can't proceed with any of the operations that are the reason I booted. These are provided by the "apt" package, which you should not remove as it is required for the normal maintenance of a Debian system. Attempting to remove it will result in the following dire warning: WARNING: The following essential packages will be removed. This should NOT be done unless you know exactly what you are doing! apt 0 upgraded, 0 newly installed, 24 to remove and 0 not upgraded. After this operation, 261 MB disk space will be freed. You are about to do something potentially harmful. To continue type in the phrase 'Yes, do as I say!' ?] I like to do all my package management manually, so I use "systemctl mask" to prevent these units from starting. I use three levels of systemd prevention: - "systemctl stop": stop a running unit. - "systemctl disable": prevents a unit from being started by default, but does not prevent a unit from being started by another unit or by some socket witchcraft. - "systemctl mask": prevents a unit from being started. Ever. The AK47 of unit prevention. When you absolutely positively have to stop a unit from starting. (Apologies to Samuel L Jackson in "Jackie Brown".) I think I used something like: systemctl mask apt-daily-upgrade.timer systemctl mask apt-daily.service systemctl mask apt-daily.timer Now "ls -al /etc/systemd/system" contains these: lrwxrwxrwx 1 root root9 May 5 2017 apt-daily-upgrade.timer -> /dev/null lrwxrwxrwx 1 root root9 Apr 15 2017 apt-daily.service -> /dev/null lrwxrwxrwx 1 root root 9 Apr 15 2017 apt-daily.timer -> /dev/null Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Unstable update ridiculousness
On 21/09/2018 07:51, Default User wrote: Using advice given, using apt-listings and aptitude full-update, l was able to update everything except the 3 packages in question, which are now "pinned", with a priority of 3000. I find apt-mark hold more convenient than pinning. I do have both cron and anacron installed, but have never used or done anything to configure them. Note: the computer in question is not on 24/7, but is normally on part of every day. Thus anacron, per the apt-listbugs man page. I mask cron and anacron because I prefer manual control of all system maintenance tasks. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Unstable update ridiculousness
Read the bug reports and decide whether any of them will hurt you: On 20/09/2018 12:54, Default User wrote: grave bugs of libtracker-sparql-2.0-0 (2.0.3-3 → 2.1.4-1) b1 - #908800 - nautilus: can't use nautilus without tracker If you use nautilus, "apt-get install tracker". serious bugs of apt (1.6.4 → 1.7.0~rc1) b2 - #909155 - apt-cache show multiple packages produces invalid output Meh. Should be fixed. No doubt it will be fixed. But until then, I do not care. serious bugs of colord (1.3.3-2 → 1.4.3-3) b3 - #908735 - colord: need to build-depend on argyll Only affects building this package, not end users. Summary: apt(1 bug), colord(1 bug), libtracker-sparql-2.0-0(1 bug) Are you sure you want to install/upgrade the above packages? [Y/n/?/...] n At this point, instead of "n", just press "Enter". Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: update problem
On 12/09/2018 14:56, Default User wrote: Arch? [...] But the wiki is good . . . (And much of it is applicable to other distributions, including Debian.) +1 for the Arch wiki. I haver never used Arch, but the Arch wiki has helped me many times. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: File with weird permissions, impossible to delete
On 12/09/2018 00:59, Greg Wooledge wrote: On Tue, Sep 11, 2018 at 01:52:15PM +0200, Pétùr wrote: I have some files, with weird permissions: # ls -la d-wS--S--T 2 1061270772 2605320832 4096 oct. 7 2412 index.html Obvious file system corruption. Unmount, fsck, re-mount, and then be prepared to restore the data from your last good backup if/when the fsck failed to repair everything. I would also use journalctl to look for any i/o errors on the affected device. What hardware and kernel and filesystem? ext4 filesystem corruption is rare and usually suggests a hardware fault. I would run a full hardware test, including CPU and RAM, and a long SMART test on the storage if it is capable. SMART self-tests will not usually detect drive data transmission errors caused by cable problems, but the SMART logs should list the CRC error count which can indicate unreliable cables or controller connectors. I would first identify the *cause* and replace all unreliable hardware before bothering to restore from backup. If you have noticed some corruption, you likely also have some unnoticed corruption. Continuing to use unreliable hardware wastes your time and endangers your data. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: sid upgrade problem
On 02/09/2018 08:33, Default User wrote: Setting up libsane1:amd64 (1.0.27-1) ... Setting up sane-utils (1.0.27-1) ... Installing new version of config file /etc/user/saned ... Installing new version of config file /etc/init.d/saned ... update-inetd: warning: cannot add service, /etc/inetd.conf does not exist It looks to me that your upgrade went fine. sane-utils depends on update-inetd for some reason, perhaps in case you are using inetd to share a local scanner via your network? Like you, I do not use inetd, do not have inetutils-inetd installed, and do not have /etc/inetd.conf . I saw the same message and took it as a harmless warning. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: sid upgrade problem
On 01/09/2018 15:14, Default User wrote: On Fri, Aug 31, 2018, 21:00 Ben Caradoc-Davies wrote: On 30/08/2018 07:18, Ben Caradoc-Davies wrote: On 30/08/2018 03:46, Default User wrote: The following packages have unmet dependencies: libsane1 : Breaks: libsane (< 1.0.27-1) but 1.0.25-4.1 is installed libsane : Depends: libsane-common (= 1.0.25-4.1) but 1.0.27-1~experimental6 is to be installed [...] I always choose "Y". I keep waiting for the situation to resolve itself, but it never does. Insight? https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=905913 sane-backends 1.0.27-1 (the source package) was just uploaded and should soon be available on sid: https://tracker.debian.org/news/984005/accepted-sane-backends-1027-1-source-into-unstable/ This upload (supposedly) fixes #905913 and so should resolve the upgrade problem. Hi, Ben. Thanks for the heads-up. I'll be watching. I just updated to 1.0.27-1 and my scanner still works (Brother MFC-L2740DW over WiFi network). Note that libsane is removed and replaced with libsane1. Package manager insistence on removing libsane is by design. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: sid upgrade problem
On 30/08/2018 07:18, Ben Caradoc-Davies wrote: On 30/08/2018 03:46, Default User wrote: The following packages have unmet dependencies: libsane1 : Breaks: libsane (< 1.0.27-1) but 1.0.25-4.1 is installed libsane : Depends: libsane-common (= 1.0.25-4.1) but 1.0.27-1~experimental6 is to be installed [...] I always choose "Y". I keep waiting for the situation to resolve itself, but it never does. Insight? https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=905913 sane-backends 1.0.27-1 (the source package) was just uploaded and should soon be available on sid: https://tracker.debian.org/news/984005/accepted-sane-backends-1027-1-source-into-unstable/ This upload (supposedly) fixes #905913 and so should resolve the upgrade problem. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: sid upgrade problem
On 30/08/2018 03:46, Default User wrote:
Running sid (Debian Unstable).
For several weeks, one particular update seems to be stalled. All other
updates work as normal:
user@domain:~$ sudo aptitude -Pvv full-upgrade
[sudo] password for default:
The following NEW packages will be installed:
libsane1{ab}
The following packages will be upgraded:
libsane-common sane-utils
2 packages upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 3,028 kB of archives. After unpacking 8,432 kB will be used.
The following packages have unmet dependencies:
libsane1 : Breaks: libsane (< 1.0.27-1) but 1.0.25-4.1 is installed
libsane : Depends: libsane-common (= 1.0.25-4.1) but
1.0.27-1~experimental6 is to be installed
The following actions will resolve these dependencies:
Keep the following packages at their current version:
1) libsane-common [1.0.25-4.1 (now, unstable)]
2) libsane1 [Not Installed]
3) sane-utils [1.0.25-4.1 (now)]
Accept this solution? [Y/n/q/?]
I always choose "Y". I keep waiting for the situation to resolve itself,
but it never does.
Insight?
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=905913
Kind regards,
--
Ben Caradoc-Davies
Director
Transient Software Limited <https://transient.nz/>
New Zealand
Re: no boot - kinda
On 25/08/2018 05:51, Glenn English wrote:
This thread on grub2 is getting really scary. I'm afraid to make
changes in the grub2 configs because an error could easily brick my
machine. And futzing with /etc/default doesn't seem to impress grub2 a
whole lot. Not concerning what I want to do, anyway.
"Brick" normally means unrecoverable. While you have rescue media, this
is unlikely. Grub also has a very flexible command line that can be used
to get a system to boot. I am not saying that you won't have some
sphincter-tightening moments, but as long as you have backups, rescue
media, and a second device for googling, you should be able to recover
from any grub changes.
I think that what I'm wanting to do is completely remove some kernels
from grub2's consideration and go back to one that works for me --
that seems to be a tricky job.
The installer and the updater send magic incantations to grub2 that
successfully add/delete kernels to the boot process. Anybody have any
idea what those might be?
How about just removing all references to the ones I don't want from
/boot and running update-grub?
You could uninstall the linux-image packages you do not want to boot and
everything should just work. Let the package manager do all the work.
Remove the meta linux-image-${arch} metapackage and no new kernel
packages will be installed. "apt-mark hold" can be used to prevent
updates to the kernel packages you have installed.
Despite its ugliness, the verbose GRUB_DEFAULT option will work, will be
stable against new kernel packages, and if you get it wrong, you can
just use the grub menu to boot a kernel and fix it. This approach will
let you keep the problematic kernels on your system so that you can
investigate why they are failing. It looks like you have the makings of
a good bug report, and it would be a shame to let this one go
unreported. You can try setting GRUB_DEFAULT. It will not hurt you. To
see the grub menu ids:
grep advanced /boot/grub/grub.cfg
GRUB_DEFAULT is just the advanced submenu id, then ">", then the
advanced menuentry id you want to boot by default. For example, on my
sid, with recovery options disabled:
$ grep advanced /boot/grub/grub.cfg
submenu 'Advanced options for Debian GNU/Linux' $menuentry_id_option
'gnulinux-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b' {
menuentry 'Debian GNU/Linux, with Linux 4.17.0-3-amd64' --class debian
--class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.17.0-3-amd64-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b' {
menuentry 'Debian GNU/Linux, with Linux 4.17.0-2-amd64' --class debian
--class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.17.0-2-amd64-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b' {
menuentry 'Debian GNU/Linux, with Linux 4.17.0-1-amd64' --class debian
--class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.17.0-1-amd64-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b' {
I could then choose to boot 4.17.0-1-amd64 by setting (single line):
GRUB_DEFAULT="gnulinux-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b>gnulinux-4.17.0-1-amd64-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b"
Note the quotes. I then run "update-grub" and I am done.
I also like "apt-mark manual" to protect kernels from "apt-get autoremove".
Kind regards,
--
Ben Caradoc-Davies
Director
Transient Software Limited <https://transient.nz/>
New Zealand
Re: mailing list vs "the futur"
On 25/08/2018 05:45, Eric S Fraga wrote: On Friday, 24 Aug 2018 at 20:15, Reco wrote: -INT_MAX. I win. -1 (wraps around so = INT_MAX) and I win! No, it is the other way around for two's complement: INT_MIN = -INT_MAX - 1 Two's complement is unbalanced. You do not know the power of the dark (negative) side. For example, for signed 32-bit integers using two's complement (C/C++ int on most current architectures and JVM int). INT_MAX = 2**31 - 1 = 2147483647 INT_MIN = -2**31 = -2147483648 https://en.wikipedia.org/wiki/Integer_(computer_science) Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: question about memtest86+
On 24/08/2018 19:05, Long Wind wrote: Sorry, Ben, i made mistake, it isn't hp dx5150, it's lenovo motherboard is 865GV-M8, it uses Dual DDR 400 memory i've used it for about 5 years, it's OK except sometimes it's noisy noise must come from fans in case Fans are a common source of case noise. Hard disk drives can be noisy. Electrical components such as coils or capacitors can also whine. since it's OK, why should i bother with memory or overheat problem? Overheating can cause instability and shorten the life of your components. Cleaning the system and cleaning, replacing, or adding fans is an inexpensive and effective way of reducing overheating. Filters can also be clogged with dust or hair. Defective memory will cause system instability and silent corruption. I would much rather have 1G of reliable memory than 1.5G of unreliable memory. Memory tests will tell you whether your 1.5G is reliable. If your computer turns itself off, there is something wrong. it has 512M x 2, now i want to change one of them to 1G, so that total is 1.5G i want to use memtester to test it, but it doesn't print progress, it seems i have to wait forever. If you do not specify a number of iterations, memtester will loop infinitely. When I run a single loop, I see progress reports, which change to "ok" for each successful test: # memtester 1G 1 memtester version 4.3.0 (64-bit) Copyright (C) 2001-2012 Charles Cazabon. Licensed under the GNU General Public License version 2 (only). pagesize is 4096 pagesizemask is 0xf000 want 1024MB (1073741824 bytes) got 1024MB (1073741824 bytes), trying mlock ...locked. Loop 1/1: Stuck Address : ok Random Value: ok Compare XOR : ok Compare SUB : ok Compare MUL : ok Compare DIV : ok Compare OR : ok Compare AND : ok Sequential Increment: ok Solid Bits : ok Block Sequential: ok Checkerboard: ok Bit Spread : ok Bit Flip: ok Walking Ones: ok Walking Zeroes : ok 8-bit Writes: ok 16-bit Writes : ok Done. This test takes 3 minutes on my i7 7700 with dual-channel DDR4 RAM. Peak package temperature was 72 C. Ambient temperature is 19 C. I have a passively cooled Streacom mini-ITX case. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: no boot - kinda
On 24/08/2018 10:41, Ben Caradoc-Davies wrote: On 24/08/2018 10:37, Ben Caradoc-Davies wrote: The new format to specify a different default kernel is horrific; it contains menu items separated with ">". You need to get the full label from your generated grub.conf because it likely contains device UUIDs: GRUB_DEFAULT="gnulinux-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b>gnulinux-4.16.0-1-amd64-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b" Actually, these are filesystem UUIDs. You can see them with "blkid". To be specific, the filesystem UUID for the root filesystem to be booted. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: no boot - kinda
On 24/08/2018 10:37, Ben Caradoc-Davies wrote: The new format to specify a different default kernel is horrific; it contains menu items separated with ">". You need to get the full label from your generated grub.conf because it likely contains device UUIDs: GRUB_DEFAULT="gnulinux-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b>gnulinux-4.16.0-1-amd64-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b" Actually, these are filesystem UUIDs. You can see them with "blkid". Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: no boot - kinda
On 24/08/2018 06:14, Glenn English wrote: I'm told that grub.cfg is a place I don't want to be. Can someone tell me how to get grub to boot the working OS, or maybe how to fix the new one? The recommended way to configure Grub 2 on Debian is to edit /etc/default/grub and run update-grub . The new format to specify a different default kernel is horrific; it contains menu items separated with ">". You need to get the full label from your generated grub.conf because it likely contains device UUIDs: GRUB_DEFAULT="gnulinux-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b>gnulinux-4.16.0-1-amd64-advanced-ed4def00-71bb-4521-a16e-9551bd762b5b" You can try adding kernel command line parameters to GRUB_CMDLINE_LINUX to fix your latest kernel. You will need to run update-grub after each change. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Bugzilla
On 24/08/2018 05:48, Timothy M Butterworth wrote: All, Is there any plans to move Debian to Bugzilla and abrt? Tim Try the debian-devel list. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: question about memtest86+
On 23/08/2018 20:26, Long Wind wrote: to Ben Caradoc-Davies: my PC is HP dx5150, i use it with that memory for more than 2 years, it's OK, no overheating. maybe memory checking cause overheating? i've installed memtester, i'll try it. Any dx5150 Front Panel LEDs lit? https://support.hp.com/au-en/document/c00868914 Does the CPU fan spin? When was it last cleaned? Do you have a cat? I still suspect overheating. Sustained CPU load from a memory tester may be enough to put it over the edge. Sensors may help you gather evidence. What is your CPU model? The dx5150 has several variants and the safe operating temperatures may vary. I note that this PC is about 13 years old. This exceeds the service life of many CPU fans. Capacitors are the next suspect (motherboard or power supply), but the symptoms suggest heat. Dust will exacerbate heat problems. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: question about memtest86+
On 23/08/2018 09:58, Long Wind wrote:
i install memtest86+ of stretch to test memory
i don't see any error msg, but after a few minutes, it shutdown my PC
does that mean my memory is bad?
i read manual of memtest86+, can't find explaination
Uncommanded shutdowns may be caused by overheating. What is your CPU
temperature under load? Your BIOS may shut your computer down if a
temperature limit is exceeded. Manufacturer? Laptop? Model? CPU? BIOS
version?
I use and recommend memtester, which can be run after a normal boot
("userspace"). I found that multiple concurrent instances of memtester
detected memory problems not found by memtest86+. Run memtester as root
so it can memlock. Because memtester runs after a normal boot, you can
use a desktop temperature monitoring tool. I use the Xfce sensors applet
polling at one second intervals, and this script:
#!/bin/bash
tmax=0
while true; do
t=$((`cat /sys/class/hwmon/hwmon2/temp1_input`/1000))
tmax=$((t>tmax?t:tmax))
echo "`date '+%T'`: $t / $tmax"
sleep 1
done
I also use thermald to limit CPU temperature (but my problem was not
temperature). I use a custom service file to reduce the polling interval
to one second.
Another useful test tool is mprime <https://www.mersenne.org/download/>
(AKA Prime95 on Windows). Be sure to disable AVX with "CpuSupportsAVX=0"
in local.txt because synthetic AVX tests can be used to generate
excessive and unrealistic amounts of heat.
Kind regards,
--
Ben Caradoc-Davies
Director
Transient Software Limited <https://transient.nz/>
New Zealand
Re: USB2 or 3 WiFi dual band adapters
On 16/08/18 05:52, Miguel A. Vallejo wrote: So, I'm also looking for a Linux friendly 802.11ac USB dongle. And I should have mentioned that the TL-WN722N is a single band (2.4 GHz) b/g/n dongle, *not* dual band. -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: USB2 or 3 WiFi dual band adapters
On 14/08/18 12:14, Ben Caradoc-Davies wrote: On 12/08/18 15:12, Ben Caradoc-Davies wrote: I am using a TP-Link TL-WN722N (ath9k_htc). I have two. Inexpensive, high-gain antenna, quite reliable despite regular hard work. Since my broadband upgrade from ADSL to gigabit fibre three days ago, I can put greater load on my WiFi, and my TL-WN722N hangs under this heavy load. Downloads or uploads of 3-4 GB files succeed only about a third of the time. Hangs require USB disconnection and reconnection. I must withdraw my recommendation. Since upgrading from linux-image-4.17.0-1-amd64 4.17.8-1 to linux-image-4.17.0-2-amd64 4.17.14-1 this morning, I have given my TL-WN722N a workout, with one 3.0 GB upload and six 4.0 GB downloads with the same servers as before, all successful, with no hangs. Perhaps the problems I encountered were related to interference that has now gone away? Or have been fixed by kernel driver improvements? Anyway, my TL-WN722N now seems quite stable under load. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: USB2 or 3 WiFi dual band adapters
On 12/08/18 15:12, Ben Caradoc-Davies wrote: I am using a TP-Link TL-WN722N (ath9k_htc). I have two. Inexpensive, high-gain antenna, quite reliable despite regular hard work. Since my broadband upgrade from ADSL to gigabit fibre three days ago, I can put greater load on my WiFi, and my TL-WN722N hangs under this heavy load. Downloads or uploads of 3-4 GB files succeed only about a third of the time. Hangs require USB disconnection and reconnection. I must withdraw my recommendation. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: USB2 or 3 WiFi dual band adapters
On 12/08/18 20:55, Curt wrote: I looked at the TP-LINK TL-WN722N on Amazon.fr suggested by another poster (whom I think was Ben if memory serves, which is increasingly not exactly the case), but had trouble understanding which of the multiple versions of the device contained the desired chipset (some commenters claimed the device they bought did not work out of the box with their Linux distribution). I guess my question would be (if I was in the business of asking questions): how can you be assured when purchasing a wireless usb dongle what chipset is lodged within the device? You must check the full version as the v2 has a different chipset (see complaints in the Amazon comments). The original is Atheros, the v2 is Realtek: Atheros: https://wikidevi.com/wiki/TP-LINK_TL-WN722N Realtek: https://wikidevi.com/wiki/TP-LINK_TL-WN722N_v2 I suggest contacting the vendor for confirmation. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: USB2 or 3 WiFi dual band adapters
On 12/08/18 13:51, Joel Wirāmu Pauling wrote: On 9 August 2018 at 03:12, tony mollica wrote: Hello. I need to find a good, reliable WiFi adapter. I have an Alfa AWUS036ACH using a RTL8812au chip and there is support but it's unreliable. Connects sometimes, mostly not. My older adapters work but they're slow but maybe that's the compromise I need to resolve. What's being used reliably? Thanks, Tony [reordered by the posting order secret police] > Basically find one that uses the ath9k Chipset. They are easily the best > supported Wifi Interface. > If you need Wireless AC then ath10k based products are useable too. > The Intel ranges are OK as clients, but are not really very Opensource. > Ath9k has the best Fully Opensource impementation out of any of the > Wireless cards. > https://wikidevi.com/wiki/Ath9k +1 for ath9k. I am using a TP-Link TL-WN722N (ath9k_htc). I have two. Inexpensive, high-gain antenna, quite reliable despite regular hard work. Inadvertent enabling of QoS while idle breaks connectivity in a small range of kernel versions: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=891060 Found in versions linux/4.15.4-1, linux/4.15.11-1, linux/4.14.17-1 Fixed in versions linux/4.16.5-1, linux/4.15.17-1 Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: update hell
On 02/08/18 13:05, Default User wrote: So, if apt-get is for non-trivial upgrades, then why not for daily use? I use it daily and for trivial upgrades. Not efficient to have multiple choices. Debian, please choose one and deprecate the others. Debian is all about multiple choices. Debian tries to include everything that meets the DFSG, from choice of init system, filesystems, servers, to desktop. Debian is inclusive. Choice reduces happiness: https://www.ted.com/talks/dan_gilbert_asks_why_are_we_happy For enhanced new-user happiness, other distributions provide more curated selections. Once users have become accustomed to a curated subset, the breadth and flexibility of Debian makes it easy to reproduce a selection found in a more limited distribution. I patiently await your hate mail. You did not mention systemd; no hate mail for you! ;-) Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: update hell
On 01/08/18 11:11, Default User wrote: I was going to read up on: apt-get dist-upgrade -V -s But then today, I was able to do a full upgrade, using: sudo aptitude -Pvv update sudo aptitude -Pvv safe-upgrade sudo aptitude -Pvv full-upgrade Finally! Now, I'm sure using: apt-get dist-upgrade -V -s would have worked as well, although I thought I read somewhere that mixing apt-get and aptitude is not a good idea. I do not know anything about that. Thanks again, Ben. And, btw . . . dpkg? apt-get? aptitude? apt? synaptic? No love for synaptic? Would Debian please just settle on one, and stick with it? They do different things at different levels and seem to play nicely together. dpkg is the lowest level and manipulates files. apt-get will download them too, from various sources, and following distribution rules. apt is a friendlier higher-level command-line interface. aptitude is an ncurses frontend, good for browsing lists of packages. synaptic is like aptitude with a GTK frontend. I only use dpkg and apt-get, but when I started with Debian, I used synaptic. I came from Red Hat / Fedora, and the only things I miss from rpm and yum are yum history (with atomic reverts) and the ability to install multiple concurrent kernel packages without the API versioning silliness of Debian (which cannot co-install both 4.17.6-2 and 4.17.8-1, for example, only one linux-image-4.17.0-1-amd64). Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: update hell
On 31/07/18 12:08, Default User wrote: As for me, I think I will just wait until the "transition" is over (or at least this part of it) before upgrading packages with dependency issues. When in doubt, do nothing. When I am in doubt, I simulate a dist-upgrade with: apt-get dist-upgrade -V -s Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Unmet dependencies in installing virtualbox-5.2
On 30/07/18 15:43, 慕 冬亮 wrote: E: Unable to correct problems, you have held broken packages. What packages are held? You can list them with "apt-mark showhold". Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: update hell
On 27/07/18 16:44, Ben Caradoc-Davies wrote: On 27/07/18 16:09, Default User wrote: Debian Unstable, x86-64 [...] $ sudo aptitude -Pvv full-upgrade [...] The following packages have unmet dependencies: libkf5xmlgui5 : Depends: qtbase-abi-5-10-0 which is a virtual package, provided by: - libqt5core5a (5.10.1+dfsg-7), but 5.11.1+dfsg-6 is to be installed [...] What to do? Wait for the Qt 5.11 transition <https://release.debian.org/transitions/html/qtbase-abi-5-11-0.html> to be completed? Even though the Qt 5.11 transition is not complete, this morning it had advanced far enough for "apt-get dist-upgrade -V" to succeed for me without package removals. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: delay before desktop loads
On 30/07/18 07:07, Frank McCormick wrote: I've been experiencing a long delay - upwards of 2 minutes before the desktop on my Debian Sid loads and displays. This happens no matter what desktop I try, Cinnamon,Mate or Icewm. [...] I enter my name and password into the lightdm logon window, the wait is upwards of 2 minutes before the desktop loads. [...] Jul 29 12:06:49 franklin kernel: random: crng init done [...] Strangely if I click the mouse on the blank desktop several times, then it will start loading, but if I don't touch it, the 2 minute delay starts This started a couple of weeks ago, before that all desktops loaded quickly. How can I debug this ? Frank, this sounds very much like a hang on an uninitialised kernel crng. Does this only happen after a fresh boot, but the hang stops if you wiggle the mouse? If so, you might be impacted by the change to make the blocking getrandom kernel syscall really block until sufficient entropy is available. Here are some reports I made of other applications that were affected by this change, with detailed analysis and fixes: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=897572 https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=899080 To identify whether this is the problem, you could patch your kernel. Here are the patches I used: https://github.com/bencaradocdavies/linux/commit/f0dfb0b7b72e38093aeaa67fe1116b409c1db3dc https://github.com/bencaradocdavies/linux/commit/19e47d7049c6ca94b98cf8c00bbeb2384a9c43b9 When I had identified the application, I used ltrace to identify the library causing the problem. What is your kernel version? Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: update hell
On 27/07/18 16:44, Ben Caradoc-Davies wrote: On 27/07/18 16:09, Default User wrote: Debian Unstable, x86-64 [...] $ sudo aptitude -Pvv full-upgrade [...] The following packages have unmet dependencies: libkf5xmlgui5 : Depends: qtbase-abi-5-10-0 which is a virtual package, provided by: - libqt5core5a (5.10.1+dfsg-7), but 5.11.1+dfsg-6 is to be installed [...] What to do? Wait for the Qt 5.11 transition <https://release.debian.org/transitions/html/qtbase-abi-5-11-0.html> to be completed? When on unstable, the transition tracker is your friend: https://release.debian.org/transitions/ More specifically, I use "apt-get dist-upgrade -V", and if apt-get wants to remove anything that I do not think it should, I reply "n", and then I use "apt-get upgrade -V" instead, so that I can upgrade packages not blocked by an in-progress transition. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: update hell
On 27/07/18 16:09, Default User wrote: Debian Unstable, x86-64 [...] $ sudo aptitude -Pvv full-upgrade [...] The following packages have unmet dependencies: libkf5xmlgui5 : Depends: qtbase-abi-5-10-0 which is a virtual package, provided by: - libqt5core5a (5.10.1+dfsg-7), but 5.11.1+dfsg-6 is to be installed [...] What to do? Wait for the Qt 5.11 transition <https://release.debian.org/transitions/html/qtbase-abi-5-11-0.html> to be completed? When on unstable, the transition tracker is your friend: https://release.debian.org/transitions/ Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: trusting .deb packages
On 25/07/18 14:35, Matthew Crews wrote: On 7/24/18 7:17 PM, Rick Thomas wrote: On Jul 24, 2018, at 2:41 PM, Matthew Crews wrote: Personally, I have a low degree of trust for Mega.nz, so caveat emptor. Why do you say that? (serious question!) Have there been reports of problems? A few reasons: 1. Kim Dotcom vs the United States. There is a strong chance Mr. Dotcom will be extradited to the United States (due to Megaupload and other criminal accusations), and will leave Mega without functional leadership. Apparently, Kim Dotcom is no longer associated with Mega: https://en.wikipedia.org/wiki/Mega_(service) 2. In light of Megaupload being shut down, there's no guarantee that Mega won't have the same fate at some point, considering the services are basically identical. How do they differ from any other cloud storage provider, other than having support for client-side encryption? Nothing stops copyright infringers from uploading encrypted blobs to Amazon S3 and sharing links and keys. Their business model, which was the basis for the closure of Megaupload, seems quite different. Was Megaupload encrypted? I do not recall that it was: https://en.wikipedia.org/wiki/Megaupload What led (allegedly) to Kim Dotcom's downfall was threatening the business model of the media-industrial complex and (allegedly) boasting about it on chat: https://en.wikipedia.org/wiki/Megaupload_legal_case The indictment is not clear cut and yet to be examined by the courts: https://en.wikipedia.org/wiki/Megaupload_legal_case#Basis_of_indictment Kim Dotcom has kept the New Zealand public entertained for years: https://en.wikipedia.org/wiki/Kim_Dotcom https://en.wikipedia.org/wiki/Internet_Party_(New_Zealand) 3. Their server software is not auditable. Even though the client-side and API are public, unless the server side is auditable, you cannot effectively trust it any better than Google Drive or Dropbox. Of course, few storage providers are, and even if they are, you should be using client-side encryption that you control, anyway. If you are using megatools, you are already using open source client-side encryption software that you control. And if you are super-paranoid, double-encrypt with "gpg --symmetric --cipher-algo AES256 --s2k-digest-algo SHA512" using a different passphrase before uploading. 4. There's always been something shady about Mega that I can't put my finger on. The whole internet is a bit shady. Perhaps Mega are behind fluoridation and chemtrails? The whole point of client-side encryption is that you do not have to trust them. Granted, these aren't the strongest reasons, but for me they are strong enough. Maybe I'm wrong though? You may be proven right, but, because it is impossible to prove a negative, you will never be proven wrong. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: trusting .deb packages
On 25/07/18 13:04, Anil Duggirala wrote: Also consider using the open-source megatools package. 1.10.0 has just been released and is expected in Debian soon. megatools 1.10.0 has just been accepted into unstable and is in the build queue. I did make a search and found this package, an older version 1.9.98-1 which seems really old, so decided to go with the .deb package. I wasnt sure if such an old package would work correctly with the service, thanks a lot, 1.9.98-1+b1 has stopped working reliably with the Mega service. This problem, related to the use of TLS for uploads and causing them to sometimes hang, has been fixed in 1.10.0: Bug#904058: megatools: megaput hangs at 100% upload https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=904058 The builds completed minutes ago. I have just installed megatools 1.10.0-1 hot off incoming. It is not yet even in unstable: https://incoming.debian.org/debian-buildd/pool/main/m/megatools/megatools_1.10.0-1_amd64.deb megadf and megals from megatools 1.10.0-1 work for me on unstable. I have not tried megaput. For 1.9.98-1, the .deb for unstable needed a +b1 binary rebuilt against libcurl4. It looks like 1.10.0 has been built against libcurl4, which is not present on stretch, so you may need a package built for stretch. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: trusting .deb packages
On 25/07/18 09:51, Ben Caradoc-Davies wrote: On 25/07/18 03:45, Anil Duggirala wrote: I am thinking about installing the Mega.nz app on my Debian Stretch installation. They provide a .deb package. Is there anything I can do to ensure this is a safe package? To know that this package will not create a security vulnerability on my system? What is the minimum security procedure to follow when installing third party provided .deb packages? Also consider using the open-source megatools package. 1.10.0 has just been released and is expected in Debian soon. megatools 1.10.0 has just been accepted into unstable and is in the build queue. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: trusting .deb packages
On 25/07/18 03:45, Anil Duggirala wrote: I am thinking about installing the Mega.nz app on my Debian Stretch installation. They provide a .deb package. Is there anything I can do to ensure this is a safe package? To know that this package will not create a security vulnerability on my system? What is the minimum security procedure to follow when installing third party provided .deb packages? Also consider using the open-source megatools package. 1.10.0 has just been released and is expected in Debian soon. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: [OT] An easier database
On 23/07/18 09:26, Leandro Noferini wrote:
Ciao a tutti,
some friends with no computing skills need to catalog some objects with
some features like components, costs and images. I gave a look to base
in libreoffice suite but it looks too difficult to use and to mantain.
Is there something easier to use in linux/debian?
Something like filemaker of years ago.
For simple lists, a Google Sheets spreadsheet enables collaborative
editing. It supports images in cells by reference with =IMAGE("URL").
How many records, and do they need relations between tables? What are
the reporting requirements and use-cases? How many users? What access
control? Input validation? Do you need the ability to roll back bad changes?
If you want a proper database, I use and recommend postgresql because it
scales from tiny to massive applications and just keeps on working
without much intervention.
Kind regards,
--
Ben Caradoc-Davies
Director
Transient Software Limited <https://transient.nz/>
New Zealand
Re: Calculator with "tapes"
On 17/07/18 04:57, Ken Heard wrote: Does Debian have a calculator package which has the equivalent of the tape produced by mechanical machines to show the entire calculation. I find such "tapes" essential when for example I am adding a long list of numbers and need to check after the addition is done to verify that all the numbers were entered correctly. Regards, Ken I use a Python interpreter in a terminal or libreoffice-calc. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
Re: Looking for ratings of all-in-one printers for Linux (Ubuntu in particular)
On 10/07/18 10:39, Dan Ritter wrote: You're both missing the main point, which is that a Brother printer with BRscript/3 is essentially a Postscript printer, and you can treat it as one. No drivers needed. And AFAIK models with BR-Script3 support also support PCL6, for which we have great CUPS/foomatic generic driver support. PostScript is lovely, but can sometimes be afflicted by crazy font substitution problems that do not seem to affect PCL6. Kind regards, -- Ben Caradoc-Davies Director Transient Software Limited <https://transient.nz/> New Zealand
