Category Archives: Administrative

Fdisk Problem

I want to back up to an older SSD, but when I open it to look at it with fdisk, I get the message: “fdisk: cannot open /dev/sdb: Invalid argument”

Any ideas?

[Update Monday morning]

I gave up on the drive, and went out and bought a new one at Best Buy.

[July 17th update]

So, I copied my Linux OS to the new drive using dd. But it won’t boot. It acts like its starting to boot, but the Fedora thing just spins ad infinitum. It won’t boot in the same desktop that I copied the OS from, or from the laptop. Any ideas?

[Bumped]

Disk-Copying issues

I’m trying to clone my SSD Linux boot drive to an on-board NVME. I use dd to copy, and it says it’s done, but when I look at the drives in Disks, the main partition on the NVME is 1G smaller than the one on the SSD ( 248 versus 247, and the NVME has a few gigs of free space), and it doesn’t recognize the partition type. I’ve tried it twice, and gotten the same result. Is there something I’m missing?

[Update a few minutes later]

Never mind.

[Update a while later]

OK, so it turns out that it did work, even though it wasn’t supposed to. When I rebooted and looked at the disks again, they were identical (except that the NVME had some free space on it, so apparently it was a little larger than the SSD). All seems to be well so far, and for the first time, I’m booting from the motherboard. So if someone tells you that you can’t clone a Linux boot drive to an NVME with dd, tell them they’re wrong.

Continuing Windows Problems

Decided to start a new post from the previous one.

I followed the instructions on how to create a Windows install drive to the letter, the files seems to have been copied cleanly, but the machine still refuses to recognize the drive as a bootable drive, after several attempts.

Now what?

[Friday-afternoon update]

I hadn’t realized that I hadn’t followed up on this, but I ended up giving up and just using Media Writer in Windows from the laptop. It went off without a hitch, except that her Windows installation wasn’t salvageable. That’s why I keep data on a different drive in both OSs. I just did a clean Windows install on the new 1T NVME that I’d installed on the motherboard (this all started when she was running out of room on the 250G SSD that Windows was running on). Anyway, all is well now.

Boot-Drive Problem

I’m trying to create a bootable 32G stick from a new drive for a Windows 10 install, but for some reason neither GParted or fdisk will allow me to use more than 4M of it. ‘fdisk F’ won’t show the other 32G. Any ideas? I can’t even figure out how to wipe it to start with a clean drive.

Here’s what happens if I just want to wipe the drive:

# dd if=/dev/zero of=/dev/sdc bs=4M iflag=nocache oflag=direct
dd: error writing ‘/dev/sdc’: No space left on device
2+0 records in
1+0 records out
4194304 bytes (4.2 MB, 4.0 MiB) copied, 0.00521806 s, 804 MB/s

IOW, it’s saying it’s writing 4M, and then it runs out of room, because it’s only seeing the first 4M.

WTF?

The problem is that I’m following the instructions here. But when I look at the disk, unlike his, which is showing the whole drive, I’m seeing that 4M of unallocated space (which isn’t actually a petition, according to fdisk) at the front, and I suspect that it’s why I can’t boot from it. And I can’t get rid of it (that is, allocate it to the rest of the drive).

[Update a while later]

I’ve tried most of the things suggested in comments, but now I’ve got a new problem. I did ignore the 4M segment, and copied the Windows boot files to the 31G partition, so in theory I have a Windows boot drive. In practice, though, the BIOS on the machine I want to repair with it refuses to see it (MSI board). Moreover, I did manage to boot into Fedora, and Nautilus in Fedora saw it, even though the BIOS doesn’t. But to top it off, Fedora is becoming unusable on the machine, freezing shortly after boot, or refusing to open any programs (including a terminal), which makes it an unuseful OS.

Other than that, things are going great.

[Update a while later]

So I restored defaults on the BIOS, and now it sees the drive. Unfortunately, when I tell it to boot from it, I get “No way, Jose.”

So I still don’t seem to have successfully created a Windows boot drive.

[Monday-morning update]

I’m now starting to wonder if I have hardware issues (in addition to the thumb-drive issue). I have an SSD that looks like it might have a Windows install on it, because I can see it on my machine (the problem is with Patricia’s, which has the same type of motherboard). But when I plug in into her machine, the BIOS refuses to see it. Furthermore, sometimes (and the most recent time in which I tried to look at it on her machine), the BIOS is unresponsive, or slow. The mouse is glitchy, and there are long (as in many seconds) delays between hitting keys and anything happening. Even furthermore, I’m having the Fedora problems described above.

I hate to replace the motherboard, both because of the cost and the PITA, but if a BIOS is flaky, not sure what else to do (other than maybe trying to flash it, which could be difficult if it’s refusing to see flash drives).

[Late-morning update]

I do see at the MSI site that there have been several BIOS updates since I bought the board, so maybe I should try that.

[Late-afternoon update]

Well, now I’m royally screwed. I’ve lost another computer, and am down to a single laptop.

Fedora had been going into a mode after I removed a flash drive from Nautilus of refusing to accept keystrokes, and there was nothing I could do to get it back except to reboot. I had successfully flashed the BIOS on Patricia’s machine, and decided to update my own. But when I tried it, unlike on Patricia’s machine, the flash wizard couldn’t see the update on the flash drive, so I had to abort. It still goes into the BIOS, and will boot Fedora, but now the login screen won’t accept keystrokes, so I can’t, you know, log in.

I know that there’s nothing wrong with the keyboard, because I’m using it right now, on the laptop.

[Tuesday-morning update]

So the keyboard works in the BIOS, and when it goes into grub, I can use the down arrows to select which kernel to boot. The problem arises when it boots into a login screen. I can select the user with the mouse, but when it comes up with the password box, it accepts no input from the keyboard. And it does this with more than one boot drive, so it’s not an installation issue. Something is happening during boot that makes it stop listening to the keyboard. I’d look at dmesg but, you know, no keyboard to ask to look at it.

I’m typing this from Patricia’s computer, which I booted from a Linux drive that I usually use on the laptop. The problem here is that it runs like molasses in an Arctic January, with long periods of non-responsiveness from any software I run, though over time this morning it’s gotten better. When I look at the system monitor, there is no process that’s using a lot of resources, so it’s a mystery why the machine is so slow.

[Bumped]

OK, a little progress. If I select rescue mode in grub, I keep the keyboard. I looked through the journal and saw lots of issues but none of them obviously related to this and apparently it’s something that doesn’t happen in rescue mode. I did modify fstab so that I can boot a different computer with it (normally, it mounts a separate drive to /home, so I can keep my data off the system drive). So I’m going to reboot this machine with it and see what happens.

[Afternoon update]

Bad news, and good news.

The bad news (at least in theory, if not practice) is that the Fedora drive from my machine won’t boot in her machine, despite the same motherboard, processor, etc. It just circles and never gets to a login screen. The reason it’s only bad in theory is that I had no plans to use it with her machine–it’s just concerning that I’m getting different behavior from the two machines.

The good news is that my machine has started to pay attention to the keyboard again, and I’m doing this update from it. Perhaps the process of going into the rescue mode straightened out whatever the issue was, but at least it’s functional again.

Now to try once again to build a Windows boot drive. I’m going to use a 60G SSD that has an old Fedora boot on it, which I won’t be using, so it’s safe to just wipe it clean and see if I can make it happen this time.

[Wednesday-morning update]

I had more keyboard issues yesterday, but I think I’ve resolved it. The keyboard and the USB port that I was using for the USB stick were sharing the same port on the motherboard, and it was probably causing a conflict. I moved the keyboard to a hub on a different port, and so far, so good. Now back to trying to create a Windows drive.

[Bumped]

The Trial

I shouldn’t discuss it here, but commenters are welcome to.

[Update on the evening of February 8th]

Here is my official statement to the press: ” I am pleased that the jury found in my favor on half of the statements at issue in this case, including finding my statement that Professor Mann engaged in data manipulation was not defamation.  In over a decade of litigation, the sanctions levied against Professor Mann dwarf the judgment against me.”

[Bumped]

[Update a few minutes later]

Here is the story at the WaPo. You have to go in pretty far to learn that the judgment against me was for only a thousand bucks. He owes me $4400 for sanctions from deposition.

[Late-evening update on the west coast]

Thoughts from closing yesterday:

Williams argued that no one giving evidence had questioned Mann’s “integrity, reliability or credibility.” This ridiculous claim drew an immediate objection, which was overruled — although the judge said he would re-read the instructions on that subject.

Then Williams argued that Steyn and Simberg had not proven that they really believed that what they wrote was true. This was of course a complete mis-statement and reversal of the actual law and instruction on the key issue of actual malice. The statement drew an immediate objection, which was sustained. As a result of the mis-statement, the judge told the jury to ignore Mr. Williams’s statement and said that he would re-read to the jury the instructions as to defamation. He did that at the close of the argument. However, it was not clear to me that the jury understood that the re-reading was intended to correct Mr. Williams’s false version of the law and burden of proof as to actual malice. They could well be confused as to this, which was clearly Williams’s intent.

Next, Williams said that his client had no need to prove monetary harm, but the jury could award damages based on reputational or emotional harm. The problem with this one is that he had no evidence to cite as to the alleged reputational or emotional harm.

And finally, Williams made a pitch for punitive damages. In his pitch, Williams started to say “These attacks on climate scientists have to stop.” Again, there was an immediate objection, again sustained. In effect, Williams was arguing for abrogation of the First Amendment as to the climate debate. I find it outrageous. But will the jury?

Apparently, they didn’t…

[Late-night update]

Professor Curry has put up her expert report, that was not allowed into evidence, on her website.

Back On The Air (Sort Of)

Momentarily.

So, I came down with a cold on Tuesday evening. I first noticed it at a U of Michigan reception at SciTech after the main reception (attendees were in high spirits after their football team had won their first national championship after a quarter of a century, unalloyed by having to unfairly share it with Nebraska; I say “unfairly” because many think that Missouri beat them but lost on a bad call). I noticed that my throat was getting a little sore, and my nose was starting to run. (Interestingly, it was the sickest I’ve been since long before Covid).

Anyway, I was better on Wednesday, and attended the conference, then spent the later afternoon and evening with my niece who lives in Orlando. I had a scheduled flight to DC on Thursday evening, but decided to go to the airport and get out earlier on standby, and arrived here late yesterday afternoon. My nose started running again and I had a rough night sleeping, and I lost my voice during the day, but I’m on the mend now. I expect I’ll be much better tomorrow, and fine for the upcoming trial next week.

But I’m spending the weekend cramming for my upcoming testimony, probably Wednesday or Thursday, as a hostile witness for the plaintiff, so probably light blogging not just this weekend, but for the next three-and-a-half weeks. So be grateful if you get any free ice cream between now and the second week of February.

Happy 2024

Let’s try to enjoy the new year for at least a day. I fear that we’re in for a rough ride.

Speaking of which, thoughts on the coming war between the macroculture and microculture.

I’ve thought about starting a Substack, but I’m not sure I want to devote enough time to it to make it worthwhile reading.

[Update a few minutes later]

The pitfalls of benevolence.