**** BEGIN LOGGING AT Tue Apr 27 02:59:56 2021 Apr 27 03:38:32 aparcar[m]: I'd ignore it. That drama has no place on the openwrt mailing list Apr 27 03:38:48 The dbus stuff is mostly working currently. Apr 27 03:55:47 aparcar[m]: is the hiding still going on? Honestly, the lack of communication has moved beyond frustrating. Apr 27 03:58:20 russell--: what did you break now? :^) Apr 27 04:29:03 i've seen these in my test bed for a while, intermittently, no idea where it's coming from Apr 27 04:29:38 i'm guessing it's openvpn execing /sbin/ip Apr 27 04:42:20 russell--: I've seen things like that intermittently, in the past, on MIPS targets. A few times I thought it could be related to some MIPS cache changes being made/changed. Apr 27 08:54:10 [ 0.000000] MIPS: machine is Buffalo WZR-600DHP Apr 27 08:54:10 [ 0.000000] SoC: Atheros AR7161 rev 2 Apr 27 08:54:21 that particular one was on mips Apr 27 09:04:06 russell--: I guessed it was on MIPS. Unfortunately, I couldn't reproduce the problem consistently, so my tests with reverting various cache changes were inconclusive. That said, almost all the instances I saw were under QEMU, during or shortly after startup. Of the two real MIPS routers I run for for 4-5 years now, can't remember seeing this problem however. Apr 27 09:05:56 guidosarducci: I'd appreciate if you could mind your attitude a bit. I am not trying to "hide", "double down" or "paper over" anything Apr 27 09:27:14 Btw, in the OpenWrt.org website says that RC1 was released in 21th April, but in the release notes there is 26th April Apr 27 09:33:19 thanks, updated Apr 27 09:33:55 we never have clear date for releases :-) there is the time of the build, but then it's often announced a few days later Apr 27 09:35:25 zorun: so 26th is the offical one now? Apr 27 09:36:03 jow: ah, I thought so? what did you have in mind? Apr 27 09:36:52 Hauke sent the email yesterday Apr 27 09:38:15 zorun: I am just updating d.o.o and wondered what to write there, all fine Apr 27 09:38:34 I can't remember how we presented RC's there in the past Apr 27 09:41:09 zorun, Hauke: https://mirror-01.infra.openwrt.org/ Apr 27 09:41:21 is that wording fine for you or any change suggestions? Apr 27 09:43:30 jow: I would just say "in the release candidate phase" to avoid updating the text for rc2 ;) otherwise looks good to me! Apr 27 09:44:27 jow: maybe you can also move 18.06 below snapshots, and remove the "will still receive limited security and bug fixes" part? Apr 27 09:46:54 zorun: in this case I'd rather directly move it to the archive Apr 27 09:47:13 no reason to single it out if its not receiving fixes anymore Apr 27 09:50:41 I have a firewall rule that looks like this in firewall.user, "iptables -t mangle -I PREROUTING -i br-lan -m mac --mac-source 00:00:00:00:00:01 -j MARK --set-mark 1". I haven't been able to find a way to make it openwrt-friendly to put it in /etc/config/firewall instead. Any ideas? Apr 27 09:54:03 barhom: which openwrt version do you run? Apr 27 09:54:12 barhom: afair you need the lastest firewall pacakge for this Apr 27 09:54:53 jow, I'm on 21.02 Apr 27 09:55:39 jow: sure, I think it's fine to remove 18.06 Apr 27 09:55:41 barhom: that should suffice Apr 27 09:56:13 barhom: https://pastebin.com/rfU19wgt Apr 27 09:56:37 barhom: oops typo, s/src_mark/set_mark/ Apr 27 09:57:43 thanks, let me give it a try Apr 27 10:01:41 jow, always helpful - thank you. Apr 27 10:25:06 after triaging there are already a few bug reports for 21.02, nice: https://bugs.openwrt.org/index.php?&project=2&do=index&reported[]=11 Apr 27 10:53:17 jow, is it possible to create a whole new "group" of firewall so that deleting them is easy? What I mean is for example "uci show firewall2". That way I would add all my special rules into firewall2.device_setmark_1, firewall2.device_setmark_2 etc. Apr 27 10:53:37 And thus I would have an easy way to remove all rules by simply doing "uci delete firewall2" Apr 27 10:55:26 barhom: no, unfortunately that is not possible Apr 27 11:05:00 Too bad, this means I have to parse through "uci show firewall" to see and remove my rules. Thanks for the info Apr 27 11:06:55 hmm, when I'm using the testing kernel on octeon, even though CONFIG_KERNEL_SLABINFO=y it is not enabled in /proc/config.gz and I'm lacking /proc/slabinfo Apr 27 11:07:18 did anyone experience something similar when using testing kernel? Apr 27 11:08:15 stintel: maybe one of our procfs debloat patches? Apr 27 11:10:13 jow: that's defined in mm/slab_common.c and the debloat patch doesn't touch that so I doubt it Apr 27 11:11:16 git grep mm/sl shows nothing so I really doubt it, I was thinking maybe a missing new dependency but it depends on SYSFS which is enabled in our generic config Apr 27 13:28:54 are uboot configs automatically generated during build based on the device dtsi? Apr 27 13:41:17 JyZyXEL: which uboot configs? Apr 27 13:42:43 russell--: these ones https://github.com/u-boot/u-boot/tree/master/include/configs Apr 27 13:44:52 JyZyXEL: which device are we talking about, because many openwrt supported devices just use the vendor's u-boot Apr 27 13:45:35 russell--: where is it defined what exactly gets used for each device? Apr 27 13:46:57 in makefiles, which will depend on which device you are working with Apr 27 13:47:01 i don't see QCA9x in there for example Apr 27 13:47:16 is there a vendor? Apr 27 13:47:25 and model number? Apr 27 13:47:26 tplink for example Apr 27 13:47:51 git grep tplink Apr 27 13:48:19 in openwrt, that is, not the u-boot tree Apr 27 13:48:25 the only supported one in uboot tree is TPLINK_WDR4300 Apr 27 13:48:49 i dont think openwrt uses uboot for that device Apr 27 13:49:03 e.g. target/linux/ath79/image/common-tp-link.mk Apr 27 13:49:10 some people might use pepe2ks uboot "fork" / mod Apr 27 13:49:31 https://github.com/pepe2k/u-boot_mod/ that one Apr 27 13:50:32 common-tp-link.mk doesn't actually define the uboot configs used for each device? Apr 27 13:51:25 fwiw, on a project recently, we managed to get modern u-boot to build for a mt7628 Apr 27 13:51:48 JyZyXEL: what's your goal? Apr 27 13:52:33 to try and modify the amount of flash memory configured for the device Apr 27 13:52:48 im just trying to understand all the places i need to modify Apr 27 13:53:16 for that old tplink device the startpoint might better be the pepe2k mod of "original" tplink uboot i think Apr 27 13:58:53 i vaguely recall writing a patch for pepe2k's u-boot and it being rejected because it already supported the bigger flash size (i was mod'ing wdr3600's with 16MB flash parts) Apr 27 13:59:15 that was a few years back and i may be misremembering Apr 27 14:07:14 oh maybe uboot actually detects the SPI flash part used and adjusts accordingly Apr 27 14:13:27 https://github.com/u-boot/u-boot/blob/master/drivers/mtd/spi/spi-nor-ids.c so it would seem Apr 27 14:19:14 so in the case of my tplink archerc6v2, i might just modify the generic-tp-link.mk, qca9563_tplink_archer-c6-v2.dts and tplink-safeloader.c with 16M addressing Apr 27 14:35:50 Ugh… https://marc.info/?l=linux-wireless&m=161952096129714&w=4 Apr 27 14:36:59 It's unlikely these cards need more than 2.5 GT/s, but still… Apr 27 14:37:49 Hi, is there any documentation on the new VLAN. My prior understanding was you needed to configure the switch section but I've recently learnt about adding _dev and specifying the VLAN at the end. Apr 27 14:38:11 E.g. uci set network.wan_eth0_2_dev.name=eth0.2.7 (so the VLAN ID would be 7) Apr 27 14:38:27 Seems a bit weird but apprntely works, is there any documentation for this? Apr 27 14:41:08 wow almost 12 years after "first" "bug" report its reason is found Apr 27 14:41:14 ~10 Apr 27 14:44:44 does chrome now do automatic translation of github ? Apr 27 15:02:20 Hauke, 21.02-rc1 changelog : "realtek" link https://openwrt.org/toh/views/toh_dev_arch-target-cpu is "login users only" visible (reported sfan5 in #openwrt) Apr 27 15:26:36 hmmhmm Apr 27 15:27:30 i'll have to see how it is after a couple days, but it looks like those random stalls i was getting with soft offload on ramips 5.4 aren't happening in the testing kernel/5.10 Apr 27 15:28:03 *connection stalls Apr 27 15:28:11 hmm, looks like the link is broken anyway Apr 27 15:31:13 this one should work: https://openwrt.org/toh/views/toh_standard_all?dataflt[0]=target_%3Drealtek Apr 27 15:37:52 Namidairo: Never had any of those, I think. What were the symptoms? Apr 27 18:09:55 liblua5.35.3 - 5.3.5-4 - This package contains the Lua shared libraries, needed by other programs. Apr 27 18:09:58 liblua5.35.3 Apr 27 18:10:01 is that a mistake? Apr 27 18:10:48 'opkg install liblua5.3' manages to install it somehow Apr 27 18:11:45 Provides: liblua5.3 Apr 27 19:09:05 what's the actual question? Apr 27 20:15:50 karlp, oh, sorry if that was not clear. why does the package name contain '5.35.3' instead of just '5.3'? Apr 27 20:35:30 someone remind me, if i want to backup the config created with makemenu where and what file is it Apr 27 20:36:25 make menuconfig? i believe that's .config Apr 27 20:38:45 oh yeah i see it now thanks Apr 27 21:12:06 I am installing logrotate with a package.. I've created the /etc/logrotate.d/ file for it. Do I have to create a /etc/crontabs/ file as well? Apr 27 21:13:01 to call the logrotate /etc/logrotate.conf? Apr 27 21:57:44 jow: yes 26th is the offical date now, do we have a rule what step defines the release date? ;-) Apr 27 22:05:40 zorun: thanks for fixing the wiki page Apr 27 22:13:58 Habbie: maybe some ABI Version suffixing? Apr 27 22:46:48 What's up with https://openwrt.org/? Page temporarily unavailable, here… Apr 27 23:55:19 rsalvaterra: works for me right now Apr 28 02:15:55 Build [#85](https://buildbot.openwrt.org/master/images/#builders/3/builds/85) of `at91/sam9x` failed. Apr 28 02:38:55 rsalvaterra: I'd be streaming video and then one of the requests will either timeout or take 20 seconds until first byte. **** ENDING LOGGING AT Wed Apr 28 02:59:56 2021