Turris OS 3.11.7 is released!

Dear Turris users.

We are releasing Turris OS 3.11.7 out from RC branch to all of you. It means that it will be for you available in a few seconds. We appreciate anyone who tried it and provides us feedback, which helps us to improve it.

In this release, there are security fixes for bind and lighttpd and as well some package updates. Owners of Turris 1.x, there will find updated unbound to version 1.9.3, which bring bug fixes and new -V option, which prints build config. Someone can appreciate updated wireguard, nextcloud and so on.

The full changelog for this release:
• unbound, haveged, youtube-dl, wireguard, nextcloud, kernel: update
• bind, lighttpd: security update
• ucollect: CA certificate update

If anybody gets in past downtime notification about not sending data, it should be fixed for you in this release, if you are not using approvals and you shouldn’t get the email on Wednesday about it. When you use approvals, check your Updater tab and after the approval, you should not get the email after 2 days.

Enjoy the release. If you are encounter any issues with this release, please let us know.

1 Like

It seems good so far. I didn’t have to copy over my modified luci_statistics files, so that’s good.

However, when I go to Pakon, I get an endlessly running “Loading data …” circle and CPU usage of “/usr/bin/foris” goes up to around 50% (plus or minus).

Afterward, I can no longer get to the Foris page. I noticed my browser attempts to reload the page repeatedly for some reason. I’ve got YaMon and Netdata installed. LuCI, YaMon and Netdata are not affected as far as I can tell.

I can kill -9 the foris process and it returns to normal. Let me know if there’s a flag I can enable to help track this down.

In the process of trying to do some basic troublshooting, I stopped and started /etc/init.d/foris-controller and /etc/init.d/foris-ws. I found some messages in the log repeated a few times:
foris-controller[17190]: WARNING:turrishw:unsupported TOS version (on omnia): 3

Probably nothing, but I thought it was odd.

Turris OS 3.11.7 (Omnia) and Turris OS 4.0 beta11 HBT (MOX) can´t still work together (Remote Access & Managed devices in Foris). How much longer?

Transition to 3.11.7 on my Omnia with system freshly transfered on SSD mSATA disk …

without any apparent problems - interface Foris and Luci, Pakon, Cloud, Netmeter, Storage, graphs, AdBlock, SQM QoS - all is ready.

schnapps - also no problem in the new storage settings

And mainly data collection ucollect works :slight_smile:

I tried Pakon again today and it works fine.

I have server ucollect offline

image

2019-09-17 23:00:01 err server_uplink[]: Failed to download contract status
2019-09-17 23:00:03 info nikola[]: (v43.1) recognized WAN interfaces: eth1, lo
2019-09-17 23:00:15 err turris-firewall-rules[]: (v63) Failed to download https://api.turris.cz/firewall/turris-ipsets.gz.sign
.
2019-09-17 23:02:01 err server_uplink[]: Failed to get registration code
.
2019-09-18 00:00:02 err server_uplink[]: Failed to download contract status
2019-09-18 00:00:02 info updater-supervisor[]: Suspending updater start for 5337 seconds
2019-09-18 00:00:03 info nikola[]: (v43.1) recognized WAN interfaces: eth1, lo
2019-09-18 00:00:09 err nikola[]: (v43.1) Exception thrown: failed to initialize crypto library
2019-09-18 00:00:09 err nikola[]: (v43.1) Exception traceback: [<FrameSummary file /usr/lib/python3.6/site-packages/nikola/__main__.py, line 186 in main>, <FrameSummary file /usr/lib/python3.6/site-packages/nikola/rpc_wrapper.py, line 30 in __init__>]
.
2019-09-18 02:02:01 err server_uplink[]: Failed to get registration code
2019-09-18 02:02:08 err turris-firewall-rules[]: (v63) Failed to download https://api.turris.cz/firewall/turris-ipsets.gz.sign
.
2019-09-18 05:02:01 err server_uplink[]: Failed to get registration code
.
2019-09-18 06:00:01 err server_uplink[]: Failed to download contract status
2019-09-18 06:00:02 info nikola[]: (v43.1) recognized WAN interfaces: eth1, lo
2019-09-18 06:00:08 err nikola[]: (v43.1) Exception thrown: failed to initialize crypto library
2019-09-18 06:00:08 err nikola[]: (v43.1) Exception traceback: [<FrameSummary file /usr/lib/python3.6/site-packages/nikola/__main__.py, line 186 in main>, <FrameSummary file /usr/lib/python3.6/site-packages/nikola/rpc_wrapper.py, line 30 in __init__>]

hello there. Is it possible to run this 3.11.7 on a Mox Classic?

Unfortunately not, MOX only supports TOS4 :slightly_frowning_face:.

1 Like

This log error in log on every whole hour …

2019-09-17 19:00:02 err server_uplink[]: Failed to download contract status
2019-09-17 19:00:18 err server_uplink[]: Failed to get registration code

x x x

… is - I think is caused by erroneous and conflicting timing in the file. /et/init.d/server-uplink.cron on every whole hour. Just changing the timing will bring the disappearance of errors

MAILTO=""

16 * * * * root /usr/share/server-uplink/registration_code.sh
17 * * * * root /usr/share/server-uplink/contract_valid.sh

2019-09-18 19:16:01 info /usr/sbin/cron[9291]: (root) CMD (/usr/share/server-uplink/registration_code.sh)
2019-09-18 19:16:01 info /usr/sbin/cron[9289]: (root) CMD (/usr/bin/rainbow_button_sync.sh)
2019-09-18 19:17:01 info /usr/sbin/cron[9387]: (root) CMD (/usr/bin/rainbow_button_sync.sh)
2019-09-18 19:17:01 info /usr/sbin/cron[9388]: (root) CMD (/usr/share/server-uplink/contract_valid.sh)

The mere change in implementation time will complete error disappearance. The error occurred in the past and now appeared again with the version 3.11.7. Developers need to modify the timing in the configuration file - please.

lighttpd not starting after update, manual try to start fail like this:

openwrt /etc/init.d # /usr/sbin/lighttpd -D -f /etc/lighttpd/lighttpd.conf
Undefined config variable: var.home_dir
2019-09-19 20:14:58: (configfile.c.1309) source: /etc/lighttpd/modules.d/30-webdav.load line: 30 pos: 49 parser failed somehow near here: +
2019-09-19 20:14:58: (configfile.c.1309) source: /etc/lighttpd/lighttpd.conf line: 37 pos: 8 parser failed somehow near here: (EOL)

I did not changed lighttpd config as forris and luci run under it hence all my web are on separate nginx. File 30-webdav.load is from 27.may 2019 and I don’t remember I ever touched it.

After update I am receiving repeatedly this:

##### Oznámení o chybách #####

Updater failed:

[string “backend”]:1195: [string “backend”]:379: Couldn’t read status file /usr/lib/opkg/status: /usr/lib/opkg/status: I/O error

After reboot this error is not emailed to me, router looks working, but system is mounted as read only.

schnapps list
mount: wrong fs type, bad option, bad superblock on /dev/mmcblk0p1,
       missing codepage or helper program, or other error

       In some cases useful info is found in syslog - try
       dmesg | tail or so.
Can't mount root partition

Anything I can try to fix that? Before upgrade everything was OK.

it looks like there is corrupted filesystem, question is if it is HW or SW problem:

Checking filesystem on /dev/mmcblk0p1
UUID: 79af27ae-8fc3-4262-97c0-500d46da169c
[1/7] checking root items
[2/7] checking extents
parent transid verify failed on 188739584 wanted 4061848 found 4108653
parent transid verify failed on 188739584 wanted 4061848 found 4108653
Ignoring transid failure
leaf parent key incorrect 188739584
bad block 188739584
ERROR: errors found in extent allocation tree or chunk allocation
[3/7] checking free space cache
[4/7] checking fs roots
checksum verify failed on 1242947584 found 3BA25622 wanted 73C0B9F3
checksum verify failed on 1242947584 found 3BA25622 wanted 73C0B9F3
bad tree block 1242947584, bytenr mismatch, want=1242947584, have=1242947648
The following tree block(s) is corrupted in tree 490:
	tree block bytenr: 1088233472, level: 1, node key: (170175, 108, 0)
ERROR: errors found in fs roots
found 1380786183 bytes used, error(s) found
total csum bytes: 933464
total tree bytes: 466018304
total fs tree bytes: 438558720
total extent tree bytes: 25526272
btree space waste bytes: 129768429
file data blocks allocated: 7635836928
 referenced 7521488896
1 Like

I found a weird bug haunting my TO - this first showed up with 3.11.7 (2019-10-19 somewhen in the evening CEST; afterwards I directly updated to 3.11.8 because I hoped that would stop it) and persists in 3.11.8 (happened again 2019-10-21 at around 21:00 CEST). Therefore I mention it in this thread also I already updated to 3.11.8.

installed WiFi-cards:

root@Router:~# lspci
00:01.0 PCI bridge: Marvell Technology Group Ltd. Device 6820 (rev 04)
00:02.0 PCI bridge: Marvell Technology Group Ltd. Device 6820 (rev 04)
00:03.0 PCI bridge: Marvell Technology Group Ltd. Device 6820 (rev 04)
02:00.0 Network controller: Qualcomm Atheros AR9580 Wireless Network Adapter (rev 01)

I use a R11e-HnD.

At a special point of time without anything obvious wrong in systemlog or kernellog WLAN is not accessible anymore - the SSIDs can still be found, but one cannot connect to them anymore (tried with Laptop, SGS7 and iPhone7).
The only option is to do a wifi down && wifi up, after which the clients can connect again.

Is this on omnia or mox? My omnia is running fine with 3.11.8 and i use wifi a lot.

3.11.7 is only for Turris Omnia and Turris 1.x, it is not available for MOX. I updated my post - I run a TO.