Zypper dup fails due to "update-alternatives: error: no alternatives for ld"

Greetings! I am trying to update and it is failing mid-way (see below). I am not sure what to make of this. Thoughts?

Error Message

(1384/3290) Installing: openssh-clients-10.3p1-1.1.x86_64 ...............................[done]
(1385/3290) Installing: texlive-pdftex-doc-2026.226.svn77868-58.1.noarch ................[done]
(1386/3290) Installing: texlive-hyphen-hungarian-doc-2026.226.svn78069-60.1.noarch ......[done]
(1387/3290) Installing: libminizip1-x86-64-v3-1.3.1-2.1.x86_64 ..........................[done]
(1388/3290) Installing: file-5.47-2.1.x86_64 ............................................[done]
(1389/3290) Installing: libctf0-2.45-3.1.x86_64 .........................................[done]
update-alternatives: error: no alternatives for ld
error: %prein(binutils-2.45-3.1.x86_64) scriptlet failed, exit status 2
error: binutils-2.45-3.1.x86_64: install failed
error: binutils-2.45-2.3.x86_64: erase skipped
(1390/3290) Installing: binutils-2.45-3.1.x86_64 .......................................[error]
Installation of binutils-2.45-3.1.x86_64 failed:
Error: Subprocess failed. Error: RPM failed: Command exited with status 1.
Abort, retry, ignore? [a/r/i] (a): a
Warning: %posttrans and %transfiletrigger scripts are not executed when aborting!

Problem occurred during or after installation or removal of packages:
Installation has been aborted as directed.
Please see the above error message for a hint.

Sleuthing

$ which ld
/usr/bin/ld

20:54:57 2026-04-28 1004 iqgrande@anthelle ~
$ ls -l `which ld`
lrwxrwxrwx 1 root root 20 Nov  3 07:06 /usr/bin/ld -> /etc/alternatives/ld

@iqgrande Hi, looks like the system has not been updated in some time (3,290 packages)?

What is the snapshot release your updating from? cat /etc/os-release | grep VERSION_ID

The information is below. The system is in a weird state. I encountered an error (documented in Nvidia G06 and SUSE Prime Stopped Working in Wayland - #5 by OrsoBruno). So, to remedy, I used snapper to roll back and all of the snapshots had one issue or another. This is the ā€œbestā€ one so far. My goal is to get an update that is correct so my system is at a good state for me to switch from SUSE Prime to Switcheroo (per the other issue).

cat /etc/os-release | grep VERSION_ID
VERSION_ID="20260420"

I read your email a little closer… I installed a full TeXLive (hence the large package count).

I have noticed some weird things here with ld; see a comparison below. Does anyone have any ideas how I can fix this?

System with Issues

$ ldd `which ld`
        linux-vdso.so.1 (0x00007fa8ed069000)
        libbfd-2.45.0.20251103-2.so => /lib64/libbfd-2.45.0.20251103-2.so (0x00007fa8ebc00000)
        libctf.so.0 => /lib64/libctf.so.0 (0x00007fa8ec1c7000)
        libc.so.6 => /lib64/libc.so.6 (0x00007fa8eb800000)
        libz.so.1 => /lib64/glibc-hwcaps/x86-64-v3/libz.so.1.3.1 (0x00007fa8ed01d000)
        libzstd.so.1 => /lib64/glibc-hwcaps/x86-64-v3/libzstd.so.1.5.7 (0x00007fa8ec0ed000)
        libsframe.so.2 => /lib64/libsframe.so.2 (0x00007fa8ebbf6000)
        /lib64/ld-linux-x86-64.so.2 (0x00007fa8ed06b000)
        libbfd-2.45.0.20251103-3.so => not found

Other Tumbleweed System Without Issues

$ ldd `which ld`
        linux-vdso.so.1 (0x00007fec80e89000)
        libalternatives.so.1 => /lib64/libalternatives.so.1 (0x00007fec80e52000)
        libc.so.6 => /lib64/libc.so.6 (0x00007fec80c00000)
        /lib64/ld-linux-x86-64.so.2 (0x00007fec80e8b000)

@iqgrande Hi, so what does ld --version show? Your systems running ok compare to my systems…

That looks similar to my error, although mine is more about the binutils install scriptlet apparently failing. I ignored it, and the rest of the update seems to go smoothly.

@malcolmlewis for context, my system also hasn’t been updated for some time. This is my 2nd attempt, the first one failed due to lack of space, and it botched a few things, so I rolled back to a previous snapshot. Haven’t rebooted though. Will see what happens after reboot.

Perhaps a download of the binutils rpm and install manually may help? I have two Tumbleweed systems running day to day, so tend to keep them up to date. I have a couple of other test systems I update from time to time to see what happens after a few months, they seem to do fine…

So, after rebooting, for some reason the Nvidia driver isn’t recognized. They seem to install fine though. Or is it that kernel 7.0 does not work with version 580.x? I don’t think I can get any newer driver, as I am on a Pascal card.

Did you remove nomodeset after installation of the driver?

Or do you have a black screen?

I’m running into the exact same issue as iqgrande.

trying to run ld --version gives:

ld: error while loading shared libraries: libbfd-2.45.0.20251103-3.so: cannot open shared object file: No such file or directory

I’m at a bit of a loss here, being quite inexperienced with linux. This issue started after a rollback I did due to the same Nvidia issue iqgrande referenced.

I had the same problem. I fixed this by downloading the latest tumbleweed binutils rpm and installing with sudo rpm -ivh --force --noscripts --nodeps --replacepkgs ./binutils-2.45-3.1.x86_64.rpm. Then, I ran your standard zypper in -f binutils to smooth it over. I also did a rollback from nvidia problems.

I suppose my situation is similar to what @malcolmlewis pointed out as well. 10k packages to be installed(!) Granted, I have all the latex packages installed

1 Like

Thank you.

I did it with the binutils-2.45-3.2.x86_64.rpm and it looks like the issue is fixed.

Good to hear! I know this is somewhat off topic, but what graphics problems were you seeing? My screen would go black when playing videos intermittently, opengl applications were buggy, and plasma was constantly complaining about graphics resets.

It’s mostly the same, though in my case it’s generally just KWin constantly complaining about graphics resets. Screen going black happens once in a while but it’s not that frequent.

Don’t really have problem with videos, but trying play any games is an exercise in futility.

I have a 3060ti and running KDE plasma. Issue happens regardless of using wayland or x11.

Unfortunately none of my snapshots were really usable, so I ended up just updating again. I can live with this for a few days and I was planning to switch to an AMD gpu anyway, so I’m just going to move ahead with that change a little sooner than I originally planned.

1 Like

Yeah ok, I have a 3090, and I have a buddy having problems with his 2060

Right, I should’ve checked for nomodeset. I’m still on the last snapshot, will try again tonight.

No, I don’t have a black screen. The graphics driver however, was listed as llvmpipe in Info Center. Looking back, I really should’ve thought about nomodeset before rolling back, it was a hefty 8 gig download…

This did it for me. Basically, what I did was:

  1. Grab the binutils rpm package manually.
  2. Use rpm --replacepkgs to install the rpm package
  3. Reinstall the Nvidia stuff with Yast

That took care of the graphics situation. Now the drivers are detected correctly and I have hardware acceleration again. :+1:

Do we mark threads as ā€œsolvedā€ here like in StackExchange?

Check out this answer: Black screen on Nvidia after updating to 20260428 - #197 by hendersj

I ran into this problem, and rather than using options to bypass scripts or other such stuff, I just created an update-alternatives entry so the script would work properly.

That seems to have worked well here.

In answer to the question about ā€˜solved’ - there’s a checkbox icon at the bottom of each message for the person who started the topic (and for staff) that can be used to mark the solution.

Don’t forget there was a huge update just last week due to glib changes. I had 6500 updates from that.