Sorry for any issues caused by this! It is going to be fixed in the upcoming RC version.
1 Like
But does it work at all? I think your issue is different, which has @redFox according to the details, which were given by you.
Did you experience it as well or you were just scary by one post?
We will not enable it by default for now. We were discussing it and decided not to include it. There are just a few devices, which support WPA3, but the main point is which you should be asking why OpenWrt does not have it enabled and preinstalled by default. They have it optional. but on our forum, it was described how to do that. If I am not mistaken it was because some client devices will have issues connecting to the WPA3 network neither they do not connect to it even if it was set to WPA2+WPA3 mixed mode.
In the future, we would like to do a package list for WPA3. It means that in Foris/ReForis you will see in the Updater tab option to choose WPA3 support.
It does not make sense to write your issue multiple times. You should write to tech.support@turris.cz as it was described in OP. If you wrote it there, then fine. Guys will reply to you soon and ask you to more details, if they need it.
Your reported bug affects just Turris 1.x routers. We are not sure what is causing it. We didnât look at it, so far. There are created two issues about it:
No known workaround, sorry.
Yes I had the exact same problem two weeks ago when I tested HBT branch, but I since rolled back to Turris OS 4.0.5.
In that case, you need to provide any details which should help us to reproduce the issue. We were thinking that there is possibility that one guy could have some problems with HW and it needs to be sent to us, but if it affects many people it needs to be investigated before releasing but without providing any details it is not helpful either.
But does it work at all? I think your issue is different, which has @redFox according to the details, which were given by you.Actually luci_statistics now works again after a reboot - most likely it was fixed by the python3 to python linking workaround necessary for luci. I was just a bit confused by the error message and luci_statistics-opkg having a slightly different format than the existing one (single quotes instead of double quotes) so I didn't expect a missing python was the cause.
The ath9k still fails to load with the vmalloc error however, while it worked fine on TOS 3.
Error is:
[ 17.898428] vmap allocation for size 356352 failed: use vmalloc= to increase size
[ 17.906477] kmodloader: vmalloc: allocation failure: 352256 bytes, mode:0x14000c0(GFP_KERNEL), nodemask=(null)
[ 17.916514] kmodloader cpuset=/ mems_allowed=0
[ 17.920973] CPU: 1 PID: 1386 Comm: kmodloader Not tainted 4.14.179 #0
Please provide your adblock config - thanks!
1 Like
Would you please provide more details or send us diagnostics, please? In the diagnostics should be more details like if 2.4 GHz Wi-Fi card is detected by HW (lspci) and if the issues is elsewhere.
In the meantime, I created an issue from details, which were provided. Unfortunately, there are not many.
// EDIT: We have first diagnostics! Thanks.
This time, we are doing a huge exception. In the previous RC2 version, many of you noticed that UIs (Foris, LuCI) were not accessible after an update. There was a leftover in one file, which was missed and should be corrected now and if not, we added a fallback solution, which creates a symlink for python.
Other changes in the RC3 version of Turris OS 5.0:
- There is an updated kernel to version 4.14.180.
- libjson-c: fixes for CVE-2020-12762
If anything shows up, please let us know immediately, so we can take a look.
hmm, everything went smooth, but i now have this in syslogâŚ
sentinel-dynfw-client[7245]: ipset v7.3: Error in line 1: Element cannot be deleted from the set: it's not added
May 16 09:45:50 turrisMOX sentinel-dynfw-client[7245]: 2020-05-16 11:45:50,206 - WARNING - Error running ipset command: return code 1.Email with diag and ipset list send
Thanks for reporting. In one of the previous RC versions, we did a small improvement for Sentinel DynFW client and now we forward stderr to syslog by default.
I created an issue from your output:
In the meantime, would you please send me an email to tech.support@turris.cz, where you will include diagnostics and as well output from ipset list | head? This could help us to take a look at it and if we need any more details, we will get back to you soon!
1 Like
Mox Clasic TOS 5.0.0 branch HBK Kernel 4.14.180
Foris OK - all tabs working, reForis & LuCi dtto
I have find the error. I have load to many blocklist. After i reduce the blocklist, adblock load again.
MFG redFOX
Since a few weeks i have issues with adblock on this HBT os. Some IPâs ( even whitelisted ) have some sort of connection problem. If i ping that specific site, it sometimes works, and sometimes not. If i switch adblock off, no problem.
Tried diff lists, but no result.
I know, noone can do precise predictions, but are there still many issues open for 5.0.0 or will it be any time soon, when we can expect this release to be approved for hbs?
I need to rearrange my network setup and want to start with current firmwares onlyâŚ
(And going with not stable firmware has a zero at the women/family-acceptance-factor)
This time, we are releasing a new RC4 version of Turris OS 5.0, which is based on the latest OpenWrt 19.07.3 release.
This release is an important one as there is a mitigation for NXNSAttack security vulnerability. This affects many DNS resolvers. We created several pull requests for all OpenWrt versions. More details about it can be found on our blog, where colleagues from Knot Resolver described in more details.
There is updated php7, golang and added hack for ppp, which improves stability in some cases.
We are getting closer to release it to the Stable branch, but there a few things, which we need to get sorted out as reported in some cases ath9k driver is not loaded and services like transmission, umdns do not start.
If you know about anything else, please let us know.
7 Likes
I got a weird notification today in Foris:
Updater failed: Called uri_path on URI of scheme: https
It might be related to broken connection (and consequently DNS) at the moment. I donât expect itâs really a problem, perhaps just a confusing message.