Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

FS#1567 - BT Home Hub 5 ADSL connects but pppoa fails to establish link since r7050-9c409cb #6529

Closed
openwrt-bot opened this issue May 27, 2018 · 7 comments
Labels

Comments

@openwrt-bot
Copy link

ezplanet:

Supply the following if possible:

  • Device problem occurs on: BT Home Hub 5
  • Software versions: OpenWrt SNAPSHOT, r7050-9c409cb
  • Steps to reproduce configure ADSL PPPoA, ADSL will sync, but PPPoA link fails to be established. Downgrading to previous r6880-aadca03 build restores operation.

This occurs on 2 separate BT Home Hub 5 connected to 2 separate ADSL lines supplied by 2 different UK providers.

The following is printed in the logs:
May 27 17:15:51 marconi pppd[20201]: Plugin pppoatm.so loaded.
May 27 17:15:51 marconi pppd[20201]: PPPoATM plugin_init
May 27 17:15:51 marconi pppd[20201]: PPPoATM setdevname_pppoatm - SUCCESS:0.0.38
May 27 17:15:51 marconi pppd[20201]: pppd 2.4.7 started by root, uid 0
May 27 17:15:51 marconi kernel: [ 1223.855179] pppoa-wan: renamed from ppp0
May 27 17:15:51 marconi pppd[20201]: Using interface pppoa-wan
May 27 17:15:51 marconi pppd[20201]: Connect: pppoa-wan <--> 0.0.38

May 27 17:16:42 marconi pppd[24483]: Plugin pppoatm.so loaded.
May 27 17:16:42 marconi pppd[24483]: PPPoATM plugin_init
May 27 17:16:42 marconi pppd[24483]: PPPoATM setdevname_pppoatm - SUCCESS:0.0.38
May 27 17:16:42 marconi pppd[24483]: pppd 2.4.7 started by root, uid 0
May 27 17:16:42 marconi kernel: [ 1274.894456] pppoa-wan: renamed from ppp0
May 27 17:16:42 marconi pppd[24483]: Using interface pppoa-wan
May 27 17:16:42 marconi pppd[24483]: Connect: pppoa-wan <--> 0.0.38

repeatedly, but no packets appear to go through and a ppp link is not established.

logread|grep netifd

Sun May 27 17:09:40 2018 daemon.notice netifd: Interface 'wan' is now down
Sun May 27 17:09:40 2018 daemon.notice netifd: Interface 'wan' is setting up now
Sun May 27 17:10:11 2018 daemon.notice netifd: Interface 'wan' is now down
Sun May 27 17:10:11 2018 daemon.notice netifd: Interface 'wan' is setting up now
Sun May 27 17:10:23 2018 daemon.notice netifd: Interface 'wan' is now down
Sun May 27 17:10:23 2018 daemon.notice netifd: Interface 'wan' is setting up now
Sun May 27 17:10:54 2018 daemon.notice netifd: Interface 'wan' is now down
Sun May 27 17:10:54 2018 daemon.notice netifd: Interface 'wan' is setting up now
Sun May 27 17:11:06 2018 daemon.notice netifd: Interface 'wan' is now down
Sun May 27 17:11:06 2018 daemon.notice netifd: Interface 'wan' is setting up now
Sun May 27 17:11:37 2018 daemon.notice netifd: Interface 'wan' is now down
Sun May 27 17:11:37 2018 daemon.notice netifd: Interface 'wan' is setting up now
Sun May 27 17:11:50 2018 daemon.notice netifd: Interface 'wan' is now down
Sun May 27 17:11:50 2018 daemon.notice netifd: Interface 'wan' is setting up now

lsmod

act_connmark 2272 0
act_mirred 3744 0
act_skbedit 2560 0
ath 18307 4 ath9k,ath9k_common,ath9k_hw,ath10k_core
ath10k_core 260704 1 ath10k_pci
ath10k_pci 32419 0
ath9k 94206 0
ath9k_common 11789 1 ath9k
ath9k_hw 338444 2 ath9k,ath9k_common
atm 36594 75 ltq_atm_vr9,pppoatm,br2684
autofs4 25435 0
br2684 7040 0
[...]

This issue is very similar to what previously reported in FS#1247 although in that case it applied only to PPPoE links, whilst PPPoA links were unaffected in the same build. I have not tested this new build with a PPPoE link as it is not available to me a this point in time.

@openwrt-bot
Copy link
Author

kesslerm:

I can confirm this behaviour on a Buffalo WBMR-HP-G300H device. Since the switch to kernel 4.14, the device fails to establish a pppoa connection. The DSL chip appears to establish a connection to the exchange (Line State: UP [0x801: showtime_tc_sync] reported by /etc/init.d/dsl_control status), but no LCP Echo responses are received by pppd.

@openwrt-bot
Copy link
Author

dwmw2:

Can we try reverting upstream commit c55fa3cccbc2c672e7f118be8f7484e53a8e9e77

@openwrt-bot
Copy link
Author

dwmw2:

And do we see this on any non-Lantiq devices?

@openwrt-bot
Copy link
Author

kesslerm:

Looks like the lantiq specific patch to net/atm/common.c (0004-MIPS-lantiq-add-atm-hack.patch) has not been updated correctly to take the changes of c55fa3cccbc2c672e7f118be8f7484e53a8e9e77 into account. It currently reads :

+++ b/net/atm/common.c @@ -62,10 +62,16 @@ static void vcc_remove_socket(struct soc write_unlock_irq(&vcc_sklist_lock); }

+struct sk_buff* (*ifx_atm_alloc_tx)(struct atm_vcc *, unsigned int) = NULL;
+EXPORT_SYMBOL(ifx_atm_alloc_tx);
+
static bool vcc_tx_ready(struct atm_vcc *vcc, unsigned int size)
{
struct sock *sk = sk_atm(vcc);

  •   if (ifx_atm_alloc_tx != NULL)
    
  •           return ifx_atm_alloc_tx(vcc, size);
    
  •   if (sk_wmem_alloc_get(sk) && !atm_may_send(vcc, size)) {
              pr_debug("Sorry: wmem_alloc = %d, size = %d, sndbuf = %d\n",
                       sk_wmem_alloc_get(sk), size, sk->sk_sndbuf);
    

I suspect that the call to ifx_atm_alloc_tx() has to be moved out of vcc_tx_ready() into vcc_sendmsg().

@openwrt-bot
Copy link
Author

None:

Now that's I've my UK based ASDL test line installed II can confirm that hh5a PPPoA is broken. As a workaround/hack creating an ATM LLC bridge and running PPPoE over that interface obtains a DHCP lease but this is less than ideal due to overheads and some ISPs don't grok PPPoEoA. My syslog in non working state:

Mon Jun 11 16:27:33 2018 daemon.notice netifd: Interface 'wan' is setting up now
Mon Jun 11 16:27:33 2018 daemon.err insmod: module is already loaded - slhc
Mon Jun 11 16:27:33 2018 daemon.err insmod: module is already loaded - ppp_generic
Mon Jun 11 16:27:33 2018 daemon.err insmod: module is already loaded - pppox
Mon Jun 11 16:27:33 2018 daemon.err insmod: module is already loaded - pppoatm
Mon Jun 11 16:27:33 2018 daemon.info pppd[31532]: Plugin pppoatm.so loaded.
Mon Jun 11 16:27:33 2018 daemon.info pppd[31532]: PPPoATM plugin_init
Mon Jun 11 16:27:33 2018 daemon.info pppd[31532]: PPPoATM setdevname_pppoatm - SUCCESS:0.0.38
Mon Jun 11 16:27:33 2018 daemon.notice pppd[31532]: pppd 2.4.7 started by root, uid 0
Mon Jun 11 16:27:33 2018 kern.info kernel: [ 9537.973467] pppoa-wan: renamed from ppp0
Mon Jun 11 16:27:33 2018 daemon.info pppd[31532]: Using interface pppoa-wan
Mon Jun 11 16:27:33 2018 daemon.notice pppd[31532]: Connect: pppoa-wan <--> 0.0.38
Mon Jun 11 16:27:36 2018 user.info adblock-[3.5.3]: adblock is currently disabled, please set adb_enabled to '1' to use this service
Mon Jun 11 16:28:03 2018 daemon.info pppd[31532]: No response to 5 echo-requests
Mon Jun 11 16:28:03 2018 daemon.notice pppd[31532]: Serial link appears to be disconnected.
Mon Jun 11 16:28:09 2018 daemon.notice pppd[31532]: Connection terminated.
Mon Jun 11 16:28:09 2018 daemon.notice pppd[31532]: Modem hangup
Mon Jun 11 16:28:09 2018 daemon.info pppd[31532]: Exit.

So far, I know, I'm not reporting anything new other than a workaround. I DO have a test ADSL line, a hh5a and the ability to build images to test patches.... so fire away :-)

Minor edit: Just proved that commit 0678cc8 (one before the bump to kernel 4.14 on lantiq) works ok. I'm building the next commit along.... I half expect it to fail but we shall see. It failed - so the switch to kernel 4.14(.37) is when the issue is first exposed 'publicly'.

@openwrt-bot
Copy link
Author

None:

Awaiting committing to master but if anyone interested: https://patchwork.ozlabs.org/project/openwrt/list/?series=50453

One day I'll write this up for posterity. An 'interesting' debugging session and stretched me to my limits.

@openwrt-bot
Copy link
Author

ezplanet:

Thank you Kevin. I am really grateful.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

1 participant