Re: panic: kernel diagnostic assertion "fli->fli_trans_cnt > 0"

Lloyd,

looks like you use a mix of layered file system and nfs.
Please use the command "/sbin/mount -v" before the panic to show the active mounts?
The nfs mount options are interesting too.

--
J. Hannken-Illjes - [email protected]



> On 17. Jan 2026, at 21:57, Lloyd Parkes <[email protected]> wrote:
> 
> Kia ora,
> I am running 11.99.4 built on Nov 22 on my amd64 and I can reliably panic it by building shells/bash in pkgsrc.
> 
> I have a fresh CVS checkout of pkgsrc on local disk and I have my distfiles automounted from a NetBSD 10.1 file server. It looks like the build gets as far as looking at one of the patch files and then panic. I can also reproduce this on a Xeon build box running 11.99.4 built on Dec 15.
> 
> Is this something that has been fixed? Does anyone know what is going on?
> 
> The full traceback is below.
> 
> Ngā mihi,
> Lloyd
> 
> [ 46189.552651] panic: kernel diagnostic assertion "fli->fli_trans_cnt > 0" failed: file "/home/lloyd/NetBSD/cvs/cartographer/src/sys/kern/vfs_trans.c", line 601
> [ 46189.552651] cpu0: Begin traceback...
> [ 46189.552651] vpanic() at netbsd:vpanic+0x189
> [ 46189.552651] kern_assert() at netbsd:kern_assert+0x4b
> [ 46189.562650] fstrans_done() at netbsd:fstrans_done+0xd7
> [ 46189.562650] VOP_UNLOCK() at netbsd:VOP_UNLOCK+0x88
> [ 46189.562650] nfs_lookup() at netbsd:nfs_lookup+0x696
> [ 46189.562650] layer_lookup() at netbsd:layer_lookup+0x60
> [ 46189.562650] VOP_LOOKUP() at netbsd:VOP_LOOKUP+0x83
> [ 46189.562650] getcwd_scandir() at netbsd:getcwd_scandir+0xa8
> [ 46189.562650] getcwd_common() at netbsd:getcwd_common+0x31d
> [ 46189.562650] sys_getcwd() at netbsd:sys___getcwd+0xa7
> [ 46189.562650] syscall() at netbsd:syscall+0x9d
> [ 46189.562650] --- syscall (number 296) ---
> [ 46189.562650] netbsd:syscall+0x9d:
> [ 46189.562650] cpu0: End traceback...
>