MassMesh Tech - Public Room Timeline

	MassMesh Tech	151 Members
		36 Servers

Load older messages

Sender	Message	Time
3 Jun 2021
jokeefe	I am swamped with work so probably not. New raspberry pi arrived for my other radio. Will get it setup and try an access point without internet connectivity	22:01:51
cure	ok I'm joining now	23:05:33
4 Jun 2021
Stephen304	Can anyone share their node firewall zone settings? I think my forwardings might be wrong	00:24:34
Stephen304	Nvm, it was correct :)	00:32:48
	jmaasr joined the room.	03:50:20
5 Jun 2021
	jmaasr set a profile picture.	08:55:36
cure	the library node has some f2fs corruption, specifically in the /usr/lib/lua directory. This is why ygg can not connect, it can not get lua to parse the config	20:17:47
cure	I am trying to make fsck.f2fs fix it but no luck so far.	20:24:14
Stephen304	Is the sd card itself alright?	23:06:27
cure	good question. I haven't been able to fix it yet, but I did take some notes. The situation is that the node is online, but yggdrasil can't start (and the web interface is broken) because the /usr/lib/lua directory is corrupted. That's the f2fs error: Wed Jun 2 10:00:35 2021 kern.warn kernel: [ 52.239046] F2FS-fs (loop0): inconsistent node block, nid:995, node_footer[nid:2696493969,ino:3778093658,ofs:440664256,cpver:18226067714279955680,blkaddr:836501912] Wed Jun 2 10:00:36 2021 kern.warn kernel: [ 52.264742] F2FS-fs (loop0): inconsistent node block, nid:995, node_footer[nid:2696493969,ino:3778093658,ofs:440664256,cpver:18226067714279955680,blkaddr:836501912] Wed Jun 2 10:00:36 2021 kern.warn kernel: [ 52.290430] F2FS-fs (loop0): inconsistent node block, nid:995, node_footer[nid:2696493969,ino:3778093658,ofs:440664256,cpver:18226067714279955680,blkaddr:836501912]	23:32:43
cure	on the upside, ssh works fine. Installing software still works fine, and rebooting also works fine. I added fsck.f2fs (and lsblk and blkid etc). But I don't know yet how to convince fsck.f2fs to actually do its thing; it keeps complaining it doesn't want to work on a partition that is "in use". That's despite me mounting / and /overlay ro.	23:35:31
cure	the lay of the land on that node is root@Libarynode:/usr/lib# mount /dev/root on /rom type squashfs (ro,relatime) proc on /proc type proc (rw,nosuid,nodev,noexec,noatime) sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,noatime) cgroup2 on /sys/fs/cgroup type cgroup2 (rw,nosuid,nodev,noexec,relatime,nsdelegate) tmpfs on /tmp type tmpfs (rw,nosuid,nodev,noatime) /dev/loop0 on /overlay type f2fs (ro,lazytime,noatime,background_gc=off,discard,no_heap,user_xattr,inline_xattr,inline_data,inline_dentry,extent_cache,mode=adaptive,active_logs=6,alloc_mode=reuse,fsync_mode=posix) overlayfs:/overlay on / type overlay (ro,noatime,lowerdir=/,upperdir=/overlay/upper,workdir=/overlay/work) /dev/mmcblk0p1 on /boot type vfat (rw,noatime,fmask=0022,dmask=0022,codepage=437,iocharset=iso8859-1,shortname=mixed,errors=remount-ro) tmpfs on /dev type tmpfs (rw,nosuid,relatime,size=512k,mode=755) devpts on /dev/pts type devpts (rw,nosuid,noexec,relatime,mode=600,ptmxmode=000) debugfs on /sys/kernel/debug type debugfs (rw,noatime) none on /sys/fs/bpf type bpf (rw,nosuid,nodev,noexec,noatime,mode=700) root@Libarynode:/etc# lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT loop0 7:0 0 496.8M 0 loop /overlay mmcblk0 179:0 0 14.5G 0 disk ├─mmcblk0p1 179:1 0 64M 0 part /boot └─mmcblk0p2 179:2 0 512M 0 part /rom root@Libarynode:/etc# losetup -l NAME SIZELIMIT OFFSET AUTOCLEAR RO BACK-FILE DIO LOG-SEC /dev/loop0 0 15925248 1 0 /mmcblk0p2 0 512	23:35:56
cure	I'm still a little fuzzy about the precise interaction of the partition (/dev/mmcblk0p2 -- blkid says it's a `squashfs`) partition, with overlayfs and f2fs	23:37:14
6 Jun 2021
Stephen304	I wonder if using a usb drive would make things more robust... I think the pi4 supports it out of the box now	02:15:11
Stephen304	Also is the library node a pi 4 or 3?	02:16:00
cure	it's a pi 3b	02:16:04
Stephen304	Might be good to upgrade that, I thought the 3 was pretty slow	02:17:49
Stephen304	Have you tried running the fsck from another pc with the sd card connected?	02:18:31
cure	I haven't	02:18:58
cure	I didn't actually physically touch the thing, it was all network access. Sitting a few feet away	02:19:29
@lurker42o:matrix.org	Did you take a backup of the Yggdrasil conf? Worst case we could just plop a new node down and use the same PK	13:59:09
cure	yeah I have a backup of the /etc directory of the node	14:05:38
cure	* yeah I have a backup of the /etc directory of the meshnode	14:05:51
cure	thinking about the best course of action here. taking the sd card out, doing an fsck from another machine, seems like a thing to try.	14:58:25
cure	maybe in parallel I should prepare a replacement sd card, just in case.	14:58:46
cure	I have an sd card but no pi 3b to try it from/configure it on	14:59:15
@lurker42o:matrix.org	do you have a pi4?	15:00:20
cure	yeah I do	15:01:03
@lurker42o:matrix.org	If you configured on that for now, I could pick one up with our "corporate card" and install it at the site on Thursday	15:01:19
@lurker42o:matrix.org	it might be good to get them a pi4 anyway	15:01:38

Show newer messages

Back to Room ListRoom Version: