Flint 2 (GL-MT6000 ) - bug reports - collective thread

Assuming you’ve got fast internet access, while using the Redmi AX6000 would you be able to perform a speed test from a device (e.g. an older laptop or Android phone) that’ll use 802.11n over 2.4GHz at 40MHz? And ideally you should be just a few meters away from the router.

I forgot to mention this, but it turns out the reason I didn’t experience crashes might of been because a different commit was merged and backported to 23.05. So the issue has already been fixed in OpenWrt.

I don’t know if the next GL.iNet firmware will fix the issue though, since even their test firmware isn’t fully in sync with OpenWrt.

Good to know , thanks for confirming it. It’s not a big deal really to separate the band names, my IOTs are on 2.4ghz anyways.

They already implemented a fix with 4.5.4. this is actually the bug which I originally reported at the beginning of this thread.

I believe 4.5.4 had a bug with usb-eth driver (I may be wrong), I was hitting that every so often when stressing the eth ports.

I can hit the same ssid bug on 4.5.5 release 3, just go on a facetime call and move away until the phone switches between bands, it doesn’t happen all the time but can repro.

Does it cause a kernel panic and router reboot?

I guess GL.iNet applied a different fix then, since the one from the OpenWrt repository is only 2 weeks old.

The current GL.iNet test firmware reports it’s version as r23001+722-38c150612c, which would mean that it’s 8 months old and predates the release version of OpenWrt 23.05.0-rc1. However, there’s also the “+722”, which could mean they’ve cherry picked some changes and made some changes of their own :thinking: Either way, it makes it difficult to know exactly how in sync they are with the official OpenWrt release.

1 Like

Once again:

I inserted the same SanDisk USB Flash Drive to Brume2 (GL-MT2500A) and I placed it at exactly same place of my Flint2.

Same distance and on the same angle/axis.

There is no interference to my ISP Modem/Router 2.4GHz network

The problem happen only when the USB on Flint2 is used.

Thanks

No, just hang… If there was a dump file or crash log I couldn’t find it in the system post reboot.

@JerryZhao is there a fix for this issue ?

20840.890605] ------------[ cut here ]------------
[20840.895222] NETDEV WATCHDOG: eth0 (mtk_soc_eth): transmit queue 3 timed out
[20840.902208] WARNING: CPU: 2 PID: 0 at net/sched/sch_generic.c:477 dev_watchdog+0x330/0x33c
[20840.910463] Modules linked in: pppoe ppp_async option wireguard usb_wwan rndis_host qmi_wwan pppox ppp_generic nft_fib_inet nf_flow_table_ipv6 nf_flow_table_ipv4 nf_flow_table_inet mt7915e mt76_connac_lib mt76 mac80211 libchacha20poly1305 ipt_REJECT huawei_cdc_ncm chacha_neon cfg80211 cdc_ncm cdc_ether xt_time xt_tcpudp xt_state xt_recent xt_quota xt_pkttype xt_owner xt_nat xt_multiport xt_mark xt_mac xt_limit xt_helper xt_conntrack xt_connmark xt_connlimit xt_connbytes xt_comment xt_cgroup xt_addrtype xt_TCPMSS xt_REDIRECT xt_MASQUERADE xt_LOG xt_CT usbserial usbnet slhc poly1305_neon nft_reject_ipv6 nft_reject_ipv4 nft_reject_inet nft_reject nft_redir nft_quota nft_objref nft_numgen nft_nat nft_masq nft_log nft_limit nft_hash nft_flow_offload nft_fib_ipv6 nft_fib_ipv4 nft_fib nft_ct nft_counter nft_compat nft_chain_nat nf_tables nf_reject_ipv4 nf_log_syslog nf_flow_table nf_conntrack_netlink nf_conncount mdio_netlink libcurve25519_generic libcrc32c libchacha iptable_nat iptable_mangle
[20840.910609]  iptable_filter ipheth ip_tables crc_ccitt compat cls_flower cdc_wdm cdc_acm br_netfilter arptable_filter arpt_mangle arp_tables act_vlan crypto_safexcel fuse cls_bpf act_bpf sch_tbf sch_ingress sch_htb sch_hfsc em_u32 cls_u32 cls_route cls_matchall cls_fw cls_flow cls_basic act_skbedit act_mirred act_gact xt_set ip_set_list_set ip_set_hash_netportnet ip_set_hash_netport ip_set_hash_netnet ip_set_hash_netiface ip_set_hash_net ip_set_hash_mac ip_set_hash_ipportnet ip_set_hash_ipportip ip_set_hash_ipport ip_set_hash_ipmark ip_set_hash_ipmac ip_set_hash_ip ip_set_bitmap_port ip_set_bitmap_ipmac ip_set_bitmap_ip ip_set nfnetlink ip6table_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ip6t_NPT ip6table_mangle ip6table_filter ip6_tables ip6t_REJECT x_tables nf_reject_ipv6 ifb ip6_udp_tunnel udp_tunnel tun ntfs nls_utf8 nls_iso8859_1 nls_cp437 sha1_generic seqiv md5 des_generic libdes authencesn authenc mtdoops uas usb_storage leds_gpio xhci_plat_hcd xhci_pci xhci_mtk_hcd
[20840.997935]  xhci_hcd uhci_hcd ohci_platform ohci_hcd fsl_mph_dr_of ehci_platform ehci_fsl kmwan ehci_hcd gpio_button_hotplug gl_sdk4_tertf gl_sdk4_black_white_list vfat fat exfat usbcore usb_common mii gl_sdk4_hw_info
[20841.104225] CPU: 2 PID: 0 Comm: swapper/2 Not tainted 5.15.139 #0
[20841.110302] Hardware name: GL.iNet GL-MT6000 (DT)
[20841.114989] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[20841.121929] pc : dev_watchdog+0x330/0x33c
[20841.125928] lr : dev_watchdog+0x330/0x33c
[20841.129922] sp : ffffffc008c6bdb0
[20841.133219] x29: ffffffc008c6bdb0 x28: 0000000000000140 x27: 00000000ffffffff
[20841.140335] x26: 0000000000000000 x25: 0000000000000002 x24: ffffff80008234c0
[20841.147451] x23: 0000000000000000 x22: 0000000000000001 x21: ffffffc008b26000
[20841.154565] x20: ffffff8000823000 x19: 0000000000000003 x18: ffffffc008b3f360
[20841.161681] x17: ffffffc0372a5000 x16: ffffffc008c68000 x15: 0000000000000609
[20841.168797] x14: 0000000000000203 x13: ffffffc008c6bae8 x12: ffffffc008b97360
[20841.175912] x11: 712074696d736e61 x10: ffffffc008b97360 x9 : 0000000000000000
[20841.183028] x8 : ffffffc008b3f310 x7 : ffffffc008b3f360 x6 : 0000000000000001
[20841.190143] x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000000000
[20841.197258] x2 : ffffff803fdb3090 x1 : ffffffc0372a5000 x0 : 000000000000003f
[20841.204374] Call trace:
[20841.206806]  dev_watchdog+0x330/0x33c
[20841.210454]  call_timer_fn.constprop.0+0x20/0x80
[20841.215056]  __run_timers.part.0+0x208/0x284
[20841.219309]  run_timer_softirq+0x38/0x70
[20841.223216]  _stext+0x10c/0x28c
[20841.226342]  __irq_exit_rcu+0xdc/0xfc
[20841.229992]  irq_exit+0xc/0x1c
[20841.233032]  handle_domain_irq+0x60/0x8c
[20841.236940]  gic_handle_irq+0x50/0x120
[20841.240674]  call_on_irq_stack+0x20/0x34
[20841.244582]  do_interrupt_handler+0x4c/0x54
[20841.248749]  el1_interrupt+0x2c/0x4c
[20841.252310]  el1h_64_irq_handler+0x14/0x20
[20841.256390]  el1h_64_irq+0x78/0x7c
[20841.259776]  arch_cpu_idle+0x14/0x20
[20841.263337]  do_idle+0xc0/0x150
[20841.266467]  cpu_startup_entry+0x24/0x60
[20841.270376]  secondary_start_kernel+0x130/0x140
[20841.274894]  __secondary_switched+0x50/0x54
[20841.279064] ---[ end trace 2c2182ad6920a5e1 ]---
[20841.283688] mtk_soc_eth 15100000.ethernet eth0: transmit timed out
[20841.457656] mtk_soc_eth 15100000.ethernet eth0: Link is Down
[20841.463351] mtk_soc_eth 15100000.ethernet eth1: Link is Down
[20842.110415] mtk_soc_eth 15100000.ethernet: warm reset failed
[20842.129351] mtk_soc_eth 15100000.ethernet eth0: configuring for fixed/2500base-x link mode
[20842.137740] mtk_soc_eth 15100000.ethernet eth0: Link is Up - 2.5Gbps/Full - flow control rx/tx
[20842.419044] mtk_soc_eth 15100000.ethernet eth1: PHY [mdio-bus:01] driver [RTL8221B-VB-CG 2.5Gbps PHY] (irq=62)
[20842.429280] mtk_soc_eth 15100000.ethernet eth1: configuring for phy/sgmii link mode
[20843.384651] kmwan: Delete node:wan

Do you use a managed switch/router connected to the Flint 2? Try to disable “Flow Control”.

1 Like

Hmm it seems its a issue with the ethernet driver or maybe wifi to ethernet (something hangs), though im sure OpenWrt is working on this.

Heres one issue:
kernel: mtk_eth_soc: use ADMAv1 on MT7981 and MT7986 by dangowrt · Pull Request #13858 · openwrt/openwrt · GitHub but this might not be the one you are having but it has similarities in common.

I’m really interested to see if GL-inet can branch off a newer version of OpenWrt to see if it fixes some of these issues. :smile:

The only thing i can’t figure out is why yours times out and mine not :thinking:, i see alot of people using the same ssid for 2.4ghz and 5ghz some people said in this forum that it can result in some of the issues.

1 Like

I don’t have a managed switch, just another AP (Asus AX86U) hard wired. How do you check Flow Control ? What is it for ?

Not sure, I just have a ton of devices/PCs connected. I try to use Ethernet as much as I can, but a few devices are still on WiFi. This is a great router @ $150 tbh even being kinda buggy :slight_smile:

My Asus AX86U couldn’t do wireguard with port forwarding and I spend 2x for those (2 of them), GlInet saved me for the time being until I find a better solution (probably PfSense)

2 Likes

While I may not be as advanced in using technology as some in this discussion, my recent router purchase (awaiting shipment) is primarily driven by the need to support a 2.5gb WAN, Wireguard, a user-friendly interface (I plan to assess the app as well), competitive 5GHz and 2.4GHz speeds. Given the concrete jungle scenario in Singapore with multiple households, I’m eager to see its performance.

In essence, I’m not overly concerned about peripheral features. Let’s acknowledge that, fundamentally, it’s a router designed to fulfill its intended purpose.

Wrong post body,
have any bugs to report?

@JerryZhao any predictions on when the new beta firmware will be released for testing? What bugs were resolved in this version?

One issue I have experienced, is how slow the Adguard Home on Flint 2 loads block filter lists, in some cases, I actually got the adguard home to be unresponsive I had to reboot the router.

I first loaded this one – Hagezi Pro++ (this took a long time to load)
https://raw.githubusercontent.com/hagezi/dns-blocklists/main/adblock/pro.plus.txt

Tying this Hagezi Block list TIF hangs and never loads
https://raw.githubusercontent.com/hagezi/dns-blocklists/main/adblock/tif.txt

It’s been waiting forever:

Process:
6172 1 root S 797m 81% 0% /usr/bin/AdGuardHome --glinet --no-check-update -c /etc/AdGuardHome/config.yam

Update:
I got it working by stopping adguard, going and deleting the temp / old filters from
/etc/AdGuardHome/data/filters

Then I used the Gitlab location of the blocklists and not githubusercontent locations via

1 Like

What is connected to your eth0 port? Flow control is clearly enabled. I had also problems with a crashing eth port, and disabling flow control (on my switch) helped.

Maybe try the command:

ethtool -A eth0 rx off tx off

1 Like

no new releases at the moment.

1 Like

I must say that I really dislike being insulted.

That is a plain lie. It is unusable. I did abandon 160MHz. I did everything that was suggested.
The Flint 2 just does not work.

Again, I am not a beta tester and I believe that expecting from a 200€ router to have a working wifi is not an extravagant request. Or at least getting an ETA about when the problem is going to be solved.

Now you won, I will stop coming here because this is obviously a waste of your time and mine.

3 Likes