January 2018 - Talk - lists.gtalug.org

war story: power failure makes CentOS system unbootable
by D. Hugh Redelmeier 30 Sep '22

30 Sep '22

tl;dr / spoiler: change fstab so that /boot/efi is not automatically mounted. See the recommendation at the end of this message. My gateway computer is a little PC running CentOS. It does not come back after a power failure. The reason (as best I can tell) is interesting and I think that I have a fix. My system is UEFI. It boots from a UEFI system partition, known as /boot/efi to Linux. If this gets corrupted, it won't boot. It is a VFAT partition. On my gateway (a Zotac Zbox RI321, a cute little box with two ethernet ports), the UEFI firmware apparently won't boot if the dirty bit is set in the system partition. CentOS normally runs with /boot/efi mounted. So when the system isn't shut down cleanly, the dirty bit will be on. Consequence: the system is not going to boot after a power failure. Odd observation 1: if I put a live Fedora 25 USB stick in the system and try to boot in this otherwise-unbootable state, CentOS boots from the main disk. So this really looks like a firmware bug/quirk. Odd observation 2: fsck doesn't seem to automatically fix the system partition. Once CentOS 7 booted, I dismounted /boot/efi and did an fsck on it. $ sudo fsck /dev/sda1 fsck from util-linux 2.23.2 fsck.fat 3.0.20 (12 Jun 2013) 0x25: Dirty bit is set. Fs was not properly unmounted and some data may be corrupt. 1) Remove dirty bit 2) No action ? 1 Leaving filesystem unchanged. /dev/sda1: 16 files, 2420/51145 clusters Googling got me to <https://www.centos.org/forums/viewtopic.php?t=50917> In particular, this advice seemed quite good: ==== recommendation ==== Two other changes I recommend for UEFI systems, to each OS's /etc/fstab. - For the /boot/efi mountpoint, add the mount options x-systemd.automount,noauto - Change fs_passno (last column) to a 1 or 2; the canonical fstab instructions suggest 2, but systemd treats 1 and 2 as the same. The first change means the EFI System partition will not be automatically read-write mounted at boot time; it's a bad idea this is the default because it puts the ESP at risk especially if ther are crashes, FAT has no journal and will therefore always be marked dirty in such a case; no other UEFI OS mounts the ESP by default. Second, if anything tries to access /boot/efi (read or write), systemd will automatically mount it, and because of the fs_passno 1 or 2, it will fsck it first and that fixes and clears the dirty bit in case it's set. Right now without these changes, it's just a matter of having the right number and bad timing of crashes to render the EFI System partition corrupt.

3 9

Build critique request and the story behind it.
by Russell 19 Apr '18

19 Apr '18

This year I purchased an APC UPS because of my flakey connection to the grid. Ominous read clicks coming from my boot drive cemented that decision. I am happy to say that not only did I avoid my biannual unexpected shutdowns, the UPS also sensed transient RF spikes. It begins to click, switching to the battery and back to mains, even though there is no visible disruption to service, although the emergency lighting in the hallway was triggered and beeped itself on. I decided to purchase a case and build a new desktop. I waited till now because I was loosely following Sarah Sharp, of Intel and listened to her comments on USB. Version 3.0 is RMA as of Jan this year and her general comment is that hardware development has peaked and so the immediate future is in developing software. Here is the case I chose. I take delivery tomorrow. http://www.coolermaster.com/case/lan-box-haf-series/haf-xb-evo/ Newegg has this combo on offer. Intel Core i5-7500 Kaby Lake 4-Core 3.4 GHz CPU w water cooling unit ASUS STRIX H270F GAMING MB G.SKILL 2 x 8G DDR4 As a creative anachronist or purposefull recycler, I haven't built a complete system from all new parts in a dozen years. I thought I'd get my hand in before arthritis takes over my hardware chops completely. I'd appreciate any warnings and caveats regarding this combo. I'm still window shopping but what I'm looking for is stability and longevity over gaming. This combo has DDR4 and USB 3.1. I'm kind of fixed on the usb but I'm not married to the ram. So suggestions which lower the cost are most appreciated. I have a 650w Corsair p/s which is ok but a big factor for me is the ambient noise and recommendations on a quiet supply would be appreciated. Noise is why I'm trying out water cooling. From the reviews on the case I'll probably have to replace the case fans, but who knows I might get lucky. I follow with interest the posts on UEFI, secure boot, SElinux, USB, udisks, udev, mobile linux and systemd. Now is my time to see if I have learned anything. Thanks to all posters. Regards -- Russell Sent by K-9 Mail

5 34

Any Debian Developers here can help me sign my GPG key?
by tlugys.ats＠spamgourmet.com 26 Feb '18

26 Feb '18

Hi GTALUG members, Do we have any Debian Developers here? I am in the second step of becoming New Debian Maintainer process -- https://nm.debian.org/process/194 and now need to have my GPG key signed by a Debian member. I was directed to the THUG group (Toronto (GNU)/Hurd User Group) http://www.nongnu.org/thug/ but only to find that their listed email (thug(a)gnu.org) is no longer working. So I'm in a bad position now, and really hope that there is a Debian Developer here, or you know of, that can help me. I live in GTA, North York to be exactly. Any help can comfort my desperate heart. Thanks Antonio

5 13

Linux Journal, RIP
by Stewart C. Russell 01 Feb '18

01 Feb '18

Looks like last month was their last month of even digital publication: https://www.linuxjournal.com/content/linux-journal-ceases-publication Jason Scott has saved everything he can on archive.org: https://archive.org/details/linuxjournalmagazine Looks like he could use some help creating tables of contents for all the issues, though: So, next bit. I took the table of contents from a page on the site to generate the table of contents here: https://archive.org/details/Linux-Journal-2011-06 - so, unfortunately, this has to be done for each issue. Any brave soul want to do this task? — @jscott : https://twitter.com/textfiles/status/938198975982850048 I'm sure it would be an amusing SMOP for someone on the list, and Jason's good people to work with. archive.org's Python interface is pretty powerful, too — https://internetarchive.readthedocs.io/en/latest/ cheers, Stewart

10 10

How to go fast without speculating... maybe
by David Collier-Brown 31 Jan '18

31 Jan '18

Kunle Olukotun didn't like systems that wasted their time stalled on loads and branches. He and his team at Afara Websystems therefor designed a non-speculating processor that did work without waits. It became the Sun T1. Speed without speculating The basic idea is to have more decoders than ALUs, so you can have lots of threads competing for an ALU. If, for example, thread 0 comes to a load, it will stall, so on the next instruction thread 1 gets the ALU, and runs... until it stalls and thread 2 get the ALU. Ditto for thread 3, and control goes back to thread 0, which has completed a multi-cycle fetch from cache and is ready to proceed once more. That is the basic idea of the Sun T-series processors. The strength is that the ALUs are never waiting for work. The weakness is that individual threads still have to wait for data to come from cache. You can improve on that Now imagine it isn't entire ALUs that are the available resources, its individual ALU component, like adders. Now the scenario becomes * thread 0 stalls * thread 1 get an adder * thread 2 gets a compare (really a subtracter) * thread 3 gets a branch unit, and will probably need to wait in the next cycle * thread 4 gets an adder * thread 5 gets an FPU ... and so on. Each cycle, the hardware assigns as many ALU components as it has available to threads, all of which can run. Only the stalled threads are waiting, and they don't need ALU bits to do that. Now more threads can run at the same time, the ALU components are (probabilistically) all busy, and we have increased capacity. But individual threads are still waiting for cache... Do I feel lucky? In principle, we could allocate two adders to thread 5, one doing the current instruction and another doing a subsequent, non-dependent instruction. It's not speculative, but it is out-of-order. That makes some threads twice as fast when doing non-interacting calculations. Allocate it three adders and it's three times as fast. If we're prepared to have more ALU components than decoders, decode deeply and we have enough of each to be likely to be able to find lots of non-dependent instructions, then we can be executing multiple instructions at once in multiple streams, and probabilistically get /startlingly/ better performance. I can see a new kind of optimizing compiler, too: one which tries to group non-dependent instructions together. Conclusion Is this what happens in a T5? That's a question for a hardware developer: I have no idea... yet Links: https://en.wikipedia.org/wiki/Kunle_Olukotun https://en.wikipedia.org/wiki/Afara_Websystems https://web.archive.org/web/20110720050850/http://www-hydra.stanford.edu/~k… -- David Collier-Brown, | Always do right. This will gratify System Programmer and Author | some people and astonish the rest davecb(a)spamcop.net | -- Mark Twain

8 20

Re: [GTALUG] January TASK Reminder
by David Collier-Brown 30 Jan '18

30 Jan '18

Task seems to have a permanent location now, at Cisco > <https://task.us12.list-manage.com/track/click?u=55cba2f29693cfd9f99a07a20&i…> > > > *January TASK Reminder* > > > A quick reminder this month's TASK will be held tomorrow night > (Wednesday, January 31) at our *new location* on the 29th Floor at 88 > Queens Quay West, courtesy of Cisco. > > This month's sessions will see Apurva Kumar and Jeremy Richards > discuss the newly published Dark Caracal investigation, and Lee Kagan > talk C2 tradecraft and design. See below for more details. > > As always, TASK is free to attend, no registration is required, and we > welcome you to invite friends and colleagues. > > *January TASK* > *Date: *January 31, 2018 > *Time: *6:00 pm > *Location:* > Floor 29, 88 Queens Quay West, Toronto > <https://task.us12.list-manage.com/track/click?u=55cba2f29693cfd9f99a07a20&i…> > > We look forward to seeing you there, > /TASK Steering Committee/ > > > >

1 0

Getting external IP
by William Park 30 Jan '18

30 Jan '18

Hi, I recently switched to CarryTel, and I'm getting disconnected quite often. So, I want to monitor my external IP, every minute. For this, I usually use <http://myexternalip.com/raw>, but it has rate limit. What site do you use to get your external IP? I want to round-robin through these. Google gives you this when you search "what is my ip", but it's difficult to dig it out. Simpler, the better. -- William Park <opengeometry(a)yahoo.ca>

9 15

Swap and Hibernate on SSDs
by Giles Orr 28 Jan '18

28 Jan '18

In the early days of SSDs, there was a lot of talk in the Linux community that you shouldn't have a swap partition on an SSD. An indirect side effect of that is that it becomes impossible to suspend-to-disk (aka "hibernate") on a SSD-only system. Although I suppose that you could have the swap partition and set swappiness to 0. So I have a new SSD system, and was wondering the current state of these things: - is it considered okay to have a swap partition on an SSD drive? - how should I set the swappiness? - is it considered okay to hibernate to an SSD? -- Giles https://www.gilesorr.com/ gilesorr(a)gmail.com

4 3

War Story
by Jonasz, Randall 22 Jan '18

22 Jan '18

A diegesis about my experience moving Windows 10 from a 512GB SSD to a 1TB SSD. I had EaseUS Todo Backup which promised to do the job for me. Alas when I inserted the 1TB disk into my SATA USB connector Windows 10 would not recognize the disk. So I installed the 1TB disk into my laptop and put the 512GB system disk into the adapter. All was good. The BIOS recognized both disks. I then used a bootable USB stick to load up ToDo Easeus Backup. It refused to clone the 512GB drive! What the $%@!. It said the disks were offline. :( Ok, I had enough of wasting time on Windows software. I had a bootable USB stick with Linux handy and loaded that on my laptop. Linux recognized both drives just fine. Thinking like a Linux sysadmin, I used dd command to copy sector for sector from the 512GB drive to the 1GB drive. It completed without an error. Now the moment of truth. I tried to boot Windows. Windows immediately went into recovery mode and wiped out the GPT entries for the first two partitions! Needless to say Windows did not boot. This is now 5am and I resolved not to be defeated by Microsoft! So using another laptop I created a bootable USB stick with Clonezilla; a Debian based distribution which promises to clone drives and partitions. I then booted my Windows laptop into Clonezilla. As expected it recognized both drives. I selected device to device copy with all of the default parameters. After 50 minutes Clonezilla announced the copy to be completed. Now the moment of truth. I booted Windows.....and it worked! I now had Windows 10 running off of my 1TB drive. Alas I still had one problem. Clonezilla was so thorough that it copied the disk geometry in the GPT table from the 512GB disk to the 1TB disk. Windows, being as smart as it is, thought the 1TB disk was only 512GB in size. Even though I should have known better, I tried EaseUS Partition Master to fix the GPT table. It said there was no problem with the disk$#%#$#@%! Back to Linux I went. I booted the Linux live USB again. I started gParted which immediately notified me that the GPT table was reporting the wrong size of the disk. Being a helpful Linux utility it offered to fix the problem for me. I accepted. Back to Windows 10 I went. This time Windows said the disk was 1TB in size. Yay! Now, since I use Bitlocker on my drives, when I went to resize the C drive to occupy the rest of the disk, both Windows 10 and EaseUS Partition Master said it couldn't be done. What the ^$#@! I've done this in Linux many times. Resizing encrypted partitions is not trivial but certainly doable in Linux. So after Googling I realized I had to first turn Bitlocker off. When I tried to do this, Windows 10 notified me it had to decrypt every file on my drives which could take hours. What the %$#@! I left it running and finally went to bed for four hours. When I got up, Windows 10 said Bitlocker was turned off. I then tried EaseUS Partition Master to resize the C drive. It took less than a second for it do so. I now have my 1TB SSD as my main disk in my laptop. Thank goodness Linux was available to save the day er night.A diegesis about my experience moving Windows 10 from a 512GB SSD to a 1TB SSD. I had EaseUS Todo Backup which promised to do the job for me. Alas when I inserted the 1TB disk into my SATA USB connector Windows 10 would not recognize the disk. So I installed the 1TB disk into my laptop and put the 512GB system disk into the adapter. All was good. The BIOS recognized both disks. I then used a bootable USB stick to load up ToDo Easeus Backup. It refused to clone the 512GB drive! What the $%@!. It said the disks were offline. :( Ok, I had enough of wasting time on Windows software. I had a bootable USB stick with Linux handy and loaded that on my laptop. Linux recognized both drives just fine. Thinking like a Linux sysadmin, I used dd command to copy sector for sector from the 512GB drive to the 1GB drive. It completed without an error. Now the moment of truth. I tried to boot Windows. Windows immediately went into recovery mode and wiped out the GPT entries for the first two partitions! Needless to say Windows did not boot. This is now 5am and I resolved not to be defeated by Microsoft! So using another laptop I created a bootable USB stick with Clonezilla; a Debian based distribution which promises to clone drives and partitions. I then booted my Windows laptop into Clonezilla. As expected it recognized both drives. I selected device to device copy with all of the default parameters. After 50 minutes Clonezilla announced the copy to be completed. Now the moment of truth. I booted Windows.....and it worked! I now had Windows 10 running off of my 1TB drive. Alas I still had one problem. Clonezilla was so thorough that it copied the disk geometry in the GPT table from the 512GB disk to the 1TB disk. Windows, being as smart as it is, thought the 1TB disk was only 512GB in size. Even though I should have known better, I tried EaseUS Partition Master to fix the GPT table. It said there was no problem with the disk$#%#$#@%! Back to Linux I went. I booted the Linux live USB again. I started gParted which immediately notified me that the GPT table was reporting the wrong size of the disk. Being a helpful Linux utility it offered to fix the problem for me. I accepted. Back to Windows 10 I went. This time Windows said the disk was 1TB in size. Yay! Now, since I use Bitlocker on my drives, when I went to resize the C drive to occupy the rest of the disk, both Windows 10 and EaseUS Partition Master said it couldn't be done. What the ^$#@! I've done this in Linux many times. Resizing encrypted partitions is not trivial but certainly doable in Linux. So after Googling I realized I had to first turn Bitlocker off. When I tried to do this, Windows 10 notified me it had to decrypt every file on my drives which could take hours. What the %$#@! I left it running and finally went to bed for four hours. When I got up, Windows 10 said Bitlocker was turned off. I then tried EaseUS Partition Master to resize the C drive. It took less than a second for it do so. I now have my 1TB SSD as my main disk in my laptop. Thank goodness Linux was available to save the day er night. Randy Know something about everything and everything about something --John Stuart Mill --- This email has been checked for viruses by AVG. http://www.avg.com

1 0

Carrytel.ca -- 75M/10M at $35/month
by William Park 22 Jan '18

22 Jan '18

I think I posted about Carrytel.ca, about whom I found out about from KWLUG. Today, I switched to Carrytel Cable75 (75M/10M, unlimited, $35/month). That's $25 cheaper than Teksavvy's equivalent. So far, it performs as promised. -- William Park <opengeometry(a)yahoo.ca>

1 0