Already merged into master branch:
https://github.com/accel-ppp/accel-ppp/pull/115
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
All Stories
Mar 15 2024
Mar 4 2024
Dec 6 2023
Dec 1 2023
Nov 30 2023
Nov 7 2023
Okey, so in that case, how do you deal with it?
this is not related to accel, because accel-ppp is only control-plane for pppoe sessions so slow performance of pppoe is related to kernel, not to accel-ppp
Only difference is type of accel session. ipoe vs pppoe. So this is not related to accel?
This question is not related to accel, traffic forwarding is done by kernel, not by accel-ppp so please contact netdev (https://lore.kernel.org/netdev/ ) if you want to discuss it
Nov 3 2023
Hi @Dimka88, I have implemented this feature for personal needs, but I would like to contribute with the code if possible. Please let me know how can I upload the changes so you can review. Should I do this on Github?
Oct 3 2023
Summarize:
Issue appears with veth and vlan_mon only on 4.19. If use bridge instead, all will work properly. As Deb10 support expired around 1 year we make decision to delete tests on Deb10 from github test
I can confirm that kernel 5.10.0-0.deb10.24-amd64 is fixing issue on Debian 10.
Apr 16 2023
Feb 16 2023
Jan 30 2023
Dec 22 2022
Dec 21 2022
Dec 20 2022
It is not fixed in actual version. It is required to fix here: https://github.com/accel-ppp/accel-ppp/blob/2b865db72bc2ddc6411950d72f1c23e8ef115b8a/accel-pppd/ppp/ipcp_opt_ipaddr.c#L100 , https://github.com/accel-ppp/accel-ppp/blob/2b865db72bc2ddc6411950d72f1c23e8ef115b8a/accel-pppd/ppp/ipv6cp_opt_intfid.c#L209 , https://github.com/accel-ppp/accel-ppp/blob/2b865db72bc2ddc6411950d72f1c23e8ef115b8a/accel-pppd/ctrl/ipoe/ipoe.c#L668
Dec 15 2022
Dec 8 2022
Dec 2 2022
Nov 18 2022
Attached the patch i made for the version 1.12.0-202-g28fe4de. It will use Calling Number/Called Number as RADIUS Calling-Station-Id/Called-Station-Id in case they are present in L2TP ICRQ, otherwise old beahvior is used (LAC/LNS IP addresses).
Nov 16 2022
Nov 15 2022
Oct 23 2022
UPD:// CMake Error looks pretty strange. It get info by the command git describe --tags --always
From your repo it return 6e5f998 because you clone only single branch.
root@debian11:/opt/accel-ppp/build# git describe --tags --always 6e5f998
So, maybe we have to improve generate version functionality also
Hi @v.huti , I see some cmake Errors
root@debian11:/opt/accel-ppp/build# cmake -DBUILD_IPOE_DRIVER=TRUE -DBUILD_VLAN_MON_DRIVER=TRUE -DCMAKE_INSTALL_PREFIX=/usr -DKDIR=/usr/src/linux-headers-`uname -r` -DLUA=TRUE -DCPACK_TYPE=Debian11 .. -- The C compiler identification is GNU 10.2.1 -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Check for working C compiler: /usr/bin/cc - skipped -- Detecting C compile features -- Detecting C compile features - done -- 'x86_64' CMake Error at cmake/cpack.cmake:5 (list): list index: 1 out of range (-1, 0) Call Stack (most recent call first): CMakeLists.txt:49 (include)
Oct 22 2022
Hi @Dimka88! The warnings are fixed in the following pull request:
https://github.com/accel-ppp/accel-ppp/pull/64
From my understanding, the situation is next:
In general, assuming anything about data alignment is a bad practice that leads to very obscure errors. Taking the address of a packed member is dangerous since the reduced alignment of the pointee is lost. If the pointer value is dereferenced, this can lead to memory alignment faults in some architectures.
Oct 12 2022
Oct 5 2022
I think it is not implemented yet, but is is good feature request