uci show sqm
sqm.eth1.eqdisc_opts='nat dual-srchost ack-filter'
sqm.eth1.iqdisc_opts='nat dual-dsthost ingress'
Latency: 14.73 ms (0.55 ms jitter)
Download: 21.77 Mbps (data used: 19.0 MB)
Upload: 9.73 Mbps (data used: 4.4 MB)
package is 20mb download and 10mb upload
and one more thing when i was getting stuttery movement in game. Ping plotter was showing smooth pings. no packet loss whatsever
ok, not sure if adaption->default + overhead is normal? either way... this isn't build related... and there is likely something else going on...
make a separate thread so all the sqm experts will see it. ( you can link to the post above by clicking the diagonal-chain icon on the lower right of the post )
@anon50098793 what you did? after updating the pi4. All the packages installed automatically without doing anythings.
before i used to install all the packages one by one now they are already installed.
fantastic to get feedback ( working even better ) from someone in the wild about the autorestore...
in my testing, it was just a tad unpredictable to make perma-default setting... so in future ( r145xx+ ) i'm disabling the 'apply' of this...
what this means...
- you can sysupgrade -R to perform autorestore during firstboot
- you can enable always with AUTORESTOREAPPLY=1
But I recommend manually using the method below after firstboot is all finished, so you review / verify what package actions are to be performed...
Once sysupgraded.... from luci;
- go to SYSTEM > CUSTOM COMMANDS
- click 'RUN' on 'restorepackages'
- if you are happy with the list you can type 'run' in the box, then click RUN again
From the command line (alternative to luci method);
#(or edit /autorestore.sh and run that yourself)
[root@dca632 /usbstick 41°]# restorepackages.sh
unsupported: > show
INTERACTIVE restorepackages.sh> curl [installed]
/etc/custom/autorestore.sh [default:restoredalready] add:18 rem:0
opkg install mailsend #[ok]#[ok]#customapp
opkg install libmount1 #[ok]#[ok]#customapp
opkg install mount-utils #[ok]#[ok]#customapp
opkg install libopcodes #[ok]#[ok]#customapp
opkg install libctf #[ok]#[ok]#customapp
opkg install objdump #[ok]#[ok]#customapp
opkg install binutils #[ok]#[ok]#customapp
opkg install gcc #[ok]#[ok]#customapp
opkg install make #[ok]#[ok]#customapp
opkg install git #[ok]#[ok]#customapp
opkg install boost #[ok]#[ok]#customapp
opkg install libevent2-core7 #[ok]#[ok]#customapp
opkg install libevent2-pthreads7 #[ok]#[ok]#customapp
opkg install mtr #[ok]#[ok]#customapp
opkg install libi2c #[ok]#[ok]#customapp
opkg install i2c-tools #[ok]#[ok]#customapp
opkg install kmod-i2c-core #[ok]#[ok]#customapp
opkg install kmod-i2c-bcm2835 #[ok]#[ok]#customapp
################################################ firstboot allgood 202009190702
################################### Added... 18
################################### Removed... 0
wanted to but...
pros(for me): usb boot, particularly 'boot order' aka fallback when not present etc. also tftpboot
cons: current one works pretty well ( minus the above feature ), mmc so far seems to be the quickest...
need to keep this pi as common / bugfree as possible for the builds, and i've fried the primary uart pins... and without those i'm reluctant to mess around with the bootloader...
pretty sure the openwrt images will be fine... athough the default cmdline.txt would need to be changed after writing the image, preferably to PARTUUID= should work... come to think of it... there is probably a bit of stuff in preinit / init.d / lib-upgrade... that's likely hardcoded to mmcblk... so sysupgrade would be 'broken'
( edit: not as bad as i'd guessed )
[root@dca632 /usbstick 46°]# fgrep -r mmc /lib
only a few lines tho...
i think practically the newer firmware has bugfixes particularly for usb/video... so anyone with buggy usb or video behavior would have a good practical reason to take the plunge...
next build will support an experimental 'dual firmware' feature... you'll need to run a one time script to set it up... ( which requires internet access for now )
from that point on... subsequent flashes will go to the 'alternate' partition... so you'll be able to run 20.x and keep snapshot as the second OS...
i'll extend it to support USB... but at the moment everything is tied to a common 'boot' drive/partition... (mmc)... my pi, is in a case and getting to the mmc is difficult... so i'll be looking at making that compatible with the latest pi, firmware 'boot-order'... which would allow you to have your boot files on the usb...
os1 or os2 can be anywhere ( tied to mmc for testing )... so if you wanted to have one rootfs on usb or both, this will be do-able...
if anyone is keen on this feature... i'd like to know how you would mostly use it?
- ( 20.x + snapshot ? )
- ( wifi travel mode vs home router mode? )
etc. etc. this ties in to how config 'syncing' / partition support is implemented...
another thing is rootfs size... which i've hardcoded to be the same as the baseos ( 1G )... but it is possible to make os1 and/or os2 partition sizes whatever ( i.e. half of the free sdcard space each )
( NOTE: these features also allow for optional EXCLUSIVE rootfs partition expansion... or 'live/mode'... but as I don't really need these they are not a priority )
It is my home router, so I cannot experiment much with it.
usb boot ( normal openwrt method-ish )...
[ 0.238308] 8021q: 802.1Q VLAN Support v1.8
[ 0.238548] Key type ._fscrypt registered
[ 0.238551] Key type .fscrypt registered
[ 0.241461] uart-pl011 fe201000.serial: cts_event_workaround enabled
[ 0.241509] fe201000.serial: ttyAMA0 at MMIO 0xfe201000 (irq = 14, base_baud = 0) is a PL011 rev2
[ 0.241653] fe201400.serial: ttyAMA1 at MMIO 0xfe201400 (irq = 14, base_baud = 0) is a PL011 rev2
[ 0.583413] usb 1-1: new high-speed USB device number 2 using xhci_hcd
[ 0.589044] printk: console [ttyAMA1] enabled
[ 0.807769] usb 1-1: New USB device found, idVendor=2109, idProduct=3431, bcdDevice= 4.21
[ 0.818880] bcm2835-power bcm2835-power: Broadcom BCM2835 power domains driver
[ 0.823587] usb 1-1: New USB device strings: Mfr=0, Product=1, SerialNumber=0
[ 0.823590] usb 1-1: Product: USB2.0 Hub
[ 0.848038] hub 1-1:1.0: USB hub found
[ 0.853755] mmc-bcm2835 fe300000.mmcnr: mmc_debug:0 mmc_debug2:0
[ 0.868025] hub 1-1:1.0: 4 ports detected
[ 0.872980] mmc-bcm2835 fe300000.mmcnr: DMA channel allocated
[ 1.275405] usb 1-1.1: new high-speed USB device number 3 using xhci_hcd
[ 1.326081] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[ 1.333080] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[ 1.340075] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[ 1.345577] mmc0: SDHCI controller on fe340000.emmc2 [fe340000.emmc2] using ADMA
[ 1.353572] of_cfs_init
[ 1.356065] of_cfs_init: OK
[ 1.359583] uart-pl011 fe201400.serial: no DMA platform data
[ 1.365369] Waiting for root device PARTUUID=5452574f-02...
[ 1.372879] mmc1: queuing unknown CIS tuple 0x80 (7 bytes)
[ 1.379827] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[ 1.404841] random: fast init done
[ 1.413798] usb 1-1.1: New USB device found, idVendor=0bda, idProduct=0109, bcdDevice= 9.59
[ 1.422150] usb 1-1.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 1.429452] usb 1-1.1: Product: USB2.0-CRW
[ 1.433540] usb 1-1.1: Manufacturer: Generic
[ 1.437802] usb 1-1.1: SerialNumber: 20090815198100000
[ 1.442825] mmc1: new high speed SDIO card at address 0001
[ 1.459762] usb-storage 1-1.1:1.0: USB Mass Storage device detected
[ 1.466283] scsi host0: usb-storage 1-1.1:1.0
[ 1.551424] usb 1-1.3: new high-speed USB device number 4 using xhci_hcd
[ 1.662586] usb 1-1.3: New USB device found, idVendor=05e3, idProduct=0749, bcdDevice=15.32
[ 1.670953] usb 1-1.3: New USB device strings: Mfr=3, Product=4, SerialNumber=2
[ 1.678268] usb 1-1.3: Product: USB3.0 Card Reader
[ 1.683064] usb 1-1.3: Manufacturer: Generic
[ 1.687338] usb 1-1.3: SerialNumber: 000000001532
[ 1.693790] usb-storage 1-1.3:1.0: USB Mass Storage device detected
[ 1.700187] scsi host1: usb-storage 1-1.3:1.0
[ 2.496365] scsi 0:0:0:0: Direct-Access Generic- SD/MMC 1.00 PQ: 0 ANSI: 0 CCS
[ 2.683611] usb 2-2: new SuperSpeed Gen 1 USB device number 2 using xhci_hcd
[ 2.708481] usb 2-2: New USB device found, idVendor=2357, idProduct=0601, bcdDevice=30.00
[ 2.716666] usb 2-2: New USB device strings: Mfr=1, Product=2, SerialNumber=6
[ 2.723814] usb 2-2: Product: USB 10/100/1000 LAN
[ 2.728528] usb 2-2: Manufacturer: TP-LINK
[ 2.732629] usb 2-2: SerialNumber: 000001000000
[ 2.737182] scsi 1:0:0:0: Direct-Access Generic STORAGE DEVICE 1532 PQ: 0 ANSI: 6
[ 2.948318] scsi 1:0:0:1: Direct-Access Generic STORAGE DEVICE 1532 PQ: 0 ANSI: 6
[ 2.949056] sd 1:0:0:0: [sdb] Attached SCSI removable disk
[ 3.108918] sd 0:0:0:0: [sda] 62521344 512-byte logical blocks: (32.0 GB/29.8 GiB)
[ 3.117054] sd 0:0:0:0: [sda] Write Protect is off
[ 3.122401] sd 0:0:0:0: [sda] No Caching mode page found
[ 3.127719] sd 0:0:0:0: [sda] Assuming drive cache: write through
[ 3.136536] sda: sda1 sda2 sda3
[ 3.141212] sd 0:0:0:0: [sda] Attached SCSI removable disk
[ 3.184855] sd 1:0:0:1: [sdc] 61521920 512-byte logical blocks: (31.5 GB/29.3 GiB)
[ 3.193051] sd 1:0:0:1: [sdc] Write Protect is off
[ 3.198500] sd 1:0:0:1: [sdc] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
[ 3.211010] sdc: sdc1 sdc2
[ 3.215185] sd 1:0:0:1: [sdc] Attached SCSI removable disk
[ 3.249500] EXT4-fs (sdc2): mounted filesystem without journal. Opts: (null)
[ 3.256600] VFS: Mounted root (ext4 filesystem) readonly on device 8:34.
[ 3.263808] Freeing unused kernel memory: 576K
[ 3.271539] Run /sbin/init as init process
[ 3.397002] init: Console is alive
[ 3.400511] init: - watchdog -
[ 3.719419] kmodloader: loading kernel modules from /etc/modules-boot.d/*
[ 3.747568] kmodloader: done loading kernel modules from /etc/modules-boot.d/*
[ 3.756740] init: - preinit -
[ 3.940729] bcmgenet: Skipping UMAC reset
[ 3.946148] bcmgenet fd580000.ethernet: configuring instance for external RGMII
[ 3.953629] bcmgenet fd580000.ethernet eth0: Link is Down
Press the [f] key and hit [enter] to enter failsafe mode
Press the , ,  or  key and hit [enter] to select the debug level
[ 8.007547] mount_root: mounting /dev/root
[ 8.934911] EXT4-fs (sdc2): re-mounted. Opts: (null)
- bootloader is good but updating is not exactly 'simple'
- speed seems decent on usb2 for os
- openwrt install / support needs a little tweaking...
- hold off on the bootloader update for a month or three... flash process is not foolproof ( simple enough ), and supplimental steps are not for the faint hearted...
contains updated elf+dat files within the /boot partition...
these have no effect ( are not copied ) during sysupgrade, so it should be fairly stable to sysupgrade to... however... i'll leave it labelled as testing as I messed with alot of files in between, so I don't fully trust it yet... also, persistentnlbwmon script was partly rewritten and is not finished... so you will likely not have persistent nlbwmon data on this build
if using the factory ( fac ) image... my understanding (guess) is that an old bootloader should just ignore the updated elf+dat files... but as i've updated my bootloader I was unable to test this... so I will keep labelling these builds elfdat_testing until we confirm that old bootloaders work fine with the newer elf+dat files...
should i update from
OpenWrt SNAPSHOT r14475-4a3230683b / LuCI Master git-20.259.37209-95c4082
probably not... the only reason to use the newer build is really for the mwan pr ( pre-installed ) or if your bootloader has been updated... ( but even then the older build should work as usual for mmc operations )...
daemon.err collectd: exec plugin: exec_read_one: error = Unknown qdisc type 'fq_codel' on interface 'eth1'
not using SQM...
eth1 is an USB adpater
any idea? (at least to turn it off:)
yup, this is at the bottom of /etc/config/luci_statistics and related to sqm-graphs... you can comment out those three(4) lines to turn it off...
#config collectd_exec_input 'collectd_exec_input'
# option cmduser 'nobody'
# option cmdgroup 'nogroup'
# option cmdline '/usr/libexec/collectd/sqm_collectd.sh eth1'
( and restart luci_statistics )
newer builds silence this with a one-per-boot message in /usr/libexec/collectd/sqm_collectd.sh
case "$qdisc" in
cake) handle_cake "$ifc"
mq) handle_mq "$ifc"
#ORG *) echo "Unknown qdisc type '$qdisc' on interface '$ifc'" 1>&2; ;;
if [ ! -f /tmp/cakestatwarn ]; then
echo "Unknown qdisc type '$qdisc' on interface '$ifc'" 1>&2;
i am getting the same error and i am using sqm with piece of cake
you probably need to go into the luci_statistics settings ( statistics > setup > general plugins > exec [configure] )
and update the script ( eth1 parameter ) to reference the actual interface you have sqm instantiated on ( ppp? )...
in the text box labelled 'script', you need to change 'eth1' to whatever sqm interface you are running...
currently it is pppoe-wan
Thu Oct 1 23:44:16 2020 daemon.err collectd: processes plugin: Creating taskstats handle failed.
Thu Oct 1 23:44:16 2020 daemon.err collectd: Initialization complete, entering read-loop.
Thu Oct 1 23:44:17 2020 daemon.err collectd: rrdtool plugin: rrd_update_r failed: /tmp/rrd/network/sqmcake-pppoe-wan/qdisct_latencyus-T0.rrd: opening '/tmp/rrd/network/sqmcake-pppoe-wan/qdisct_latencyus-T0.rrd': No such file or directory