| Message ID | 20240207192933.441744-1-thinker.li@gmail.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Remove expired routes with a separated list of routes. | expand |
On Wed, 2024-02-07 at 11:29 -0800, thinker.li@gmail.com wrote: > From: Kui-Feng Lee <thinker.li@gmail.com> > > This patchset is resent due to previous reverting. [1] > > FIB6 GC walks trees of fib6_tables to remove expired routes. Walking a tree > can be expensive if the number of routes in a table is big, even if most of > them are permanent. Checking routes in a separated list of routes having > expiration will avoid this potential issue. > > Background > ========== > > The size of a Linux IPv6 routing table can become a big problem if not > managed appropriately. Now, Linux has a garbage collector to remove > expired routes periodically. However, this may lead to a situation in > which the routing path is blocked for a long period due to an > excessive number of routes. > > For example, years ago, there is a commit c7bb4b89033b ("ipv6: tcp: > drop silly ICMPv6 packet too big messages"). The root cause is that > malicious ICMPv6 packets were sent back for every small packet sent to > them. These packets add routes with an expiration time that prompts > the GC to periodically check all routes in the tables, including > permanent ones. > > Why Route Expires > ================= > > Users can add IPv6 routes with an expiration time manually. However, > the Neighbor Discovery protocol may also generate routes that can > expire. For example, Router Advertisement (RA) messages may create a > default route with an expiration time. [RFC 4861] For IPv4, it is not > possible to set an expiration time for a route, and there is no RA, so > there is no need to worry about such issues. > > Create Routes with Expires > ========================== > > You can create routes with expires with the command. > > For example, > > ip -6 route add 2001:b000:591::3 via fe80::5054:ff:fe12:3457 \ > dev enp0s3 expires 30 > > The route that has been generated will be deleted automatically in 30 > seconds. > > GC of FIB6 > ========== > > The function called fib6_run_gc() is responsible for performing > garbage collection (GC) for the Linux IPv6 stack. It checks for the > expiration of every route by traversing the trees of routing > tables. The time taken to traverse a routing table increases with its > size. Holding the routing table lock during traversal is particularly > undesirable. Therefore, it is preferable to keep the lock for the > shortest possible duration. > > Solution > ======== > > The cause of the issue is keeping the routing table locked during the > traversal of large trees. To solve this problem, we can create a separate > list of routes that have expiration. This will prevent GC from checking > permanent routes. > > Result > ====== > > We conducted a test to measure the execution times of fib6_gc_timer_cb() > and observed that it enhances the GC of FIB6. During the test, we added > permanent routes with the following numbers: 1000, 3000, 6000, and > 9000. Additionally, we added a route with an expiration time. > > Here are the average execution times for the kernel without the patch. > - 120020 ns with 1000 permanent routes > - 308920 ns with 3000 ... > - 581470 ns with 6000 ... > - 855310 ns with 9000 ... > > The kernel with the patch consistently takes around 14000 ns to execute, > regardless of the number of permanent routes that are installed. > > Major changes from v4: > > - Fix the comment of fib6_add_gc_list(). > > Major changes from v3: > > - Move the checks of f6i->fib6_node to fib6_add_gc_list(). > > - Make spin_lock_bh() and spin_unlock_bh() stands out. > > - Explain the reason of the changes in the commit message of the > patch 4. > > Major changes from v2: > > - Refactory the boilerplate checks in the test case. > > - check_rt_num() and check_rt_num_clean() > > Major changes from v1: > > - Reduce the numbers of routes (5) in the test cases to work with > slow environments. Due to the failure on patchwork. > > - Remove systemd related commands in the test case. > > Major changes from the previous patchset [2]: > > - Split helpers. > > - fib6_set_expires() -> fib6_set_expires() and fib6_add_gc_list(). > > - fib6_clean_expires() -> fib6_clean_expires() and > fib6_remove_gc_list(). > > - Fix rt6_add_dflt_router() to avoid racing of setting expires. > > - Remove unnecessary calling to fib6_clean_expires() in > ip6_route_info_create(). > > - Add test cases of toggling routes between permanent and temporary > and handling routes from RA messages. > > - Clean up routes by deleting the existing device and adding a new > one. > > - Fix a potential issue in modify_prefix_route(). Note that we have a selftest failure in the batch including this series for the fib_tests: https://netdev-3.bots.linux.dev/vmksft-net/results/456022/6-fib-tests-sh/stdout I haven't digged much, but I fear its related. Please have a look. For more info on how to reproduce the selftest environment: https://github.com/linux-netdev/nipa/wiki/How-to-run-netdev-selftests-CI-style Thanks, Paolo
On 2/8/24 03:55, Paolo Abeni wrote: > On Wed, 2024-02-07 at 11:29 -0800, thinker.li@gmail.com wrote: >> From: Kui-Feng Lee <thinker.li@gmail.com> >> >> This patchset is resent due to previous reverting. [1] >> >> FIB6 GC walks trees of fib6_tables to remove expired routes. Walking a tree >> can be expensive if the number of routes in a table is big, even if most of >> them are permanent. Checking routes in a separated list of routes having >> expiration will avoid this potential issue. >> >> Background >> ========== >> >> The size of a Linux IPv6 routing table can become a big problem if not >> managed appropriately. Now, Linux has a garbage collector to remove >> expired routes periodically. However, this may lead to a situation in >> which the routing path is blocked for a long period due to an >> excessive number of routes. >> >> For example, years ago, there is a commit c7bb4b89033b ("ipv6: tcp: >> drop silly ICMPv6 packet too big messages"). The root cause is that >> malicious ICMPv6 packets were sent back for every small packet sent to >> them. These packets add routes with an expiration time that prompts >> the GC to periodically check all routes in the tables, including >> permanent ones. >> >> Why Route Expires >> ================= >> >> Users can add IPv6 routes with an expiration time manually. However, >> the Neighbor Discovery protocol may also generate routes that can >> expire. For example, Router Advertisement (RA) messages may create a >> default route with an expiration time. [RFC 4861] For IPv4, it is not >> possible to set an expiration time for a route, and there is no RA, so >> there is no need to worry about such issues. >> >> Create Routes with Expires >> ========================== >> >> You can create routes with expires with the command. >> >> For example, >> >> ip -6 route add 2001:b000:591::3 via fe80::5054:ff:fe12:3457 \ >> dev enp0s3 expires 30 >> >> The route that has been generated will be deleted automatically in 30 >> seconds. >> >> GC of FIB6 >> ========== >> >> The function called fib6_run_gc() is responsible for performing >> garbage collection (GC) for the Linux IPv6 stack. It checks for the >> expiration of every route by traversing the trees of routing >> tables. The time taken to traverse a routing table increases with its >> size. Holding the routing table lock during traversal is particularly >> undesirable. Therefore, it is preferable to keep the lock for the >> shortest possible duration. >> >> Solution >> ======== >> >> The cause of the issue is keeping the routing table locked during the >> traversal of large trees. To solve this problem, we can create a separate >> list of routes that have expiration. This will prevent GC from checking >> permanent routes. >> >> Result >> ====== >> >> We conducted a test to measure the execution times of fib6_gc_timer_cb() >> and observed that it enhances the GC of FIB6. During the test, we added >> permanent routes with the following numbers: 1000, 3000, 6000, and >> 9000. Additionally, we added a route with an expiration time. >> >> Here are the average execution times for the kernel without the patch. >> - 120020 ns with 1000 permanent routes >> - 308920 ns with 3000 ... >> - 581470 ns with 6000 ... >> - 855310 ns with 9000 ... >> >> The kernel with the patch consistently takes around 14000 ns to execute, >> regardless of the number of permanent routes that are installed. >> >> Major changes from v4: >> >> - Fix the comment of fib6_add_gc_list(). >> >> Major changes from v3: >> >> - Move the checks of f6i->fib6_node to fib6_add_gc_list(). >> >> - Make spin_lock_bh() and spin_unlock_bh() stands out. >> >> - Explain the reason of the changes in the commit message of the >> patch 4. >> >> Major changes from v2: >> >> - Refactory the boilerplate checks in the test case. >> >> - check_rt_num() and check_rt_num_clean() >> >> Major changes from v1: >> >> - Reduce the numbers of routes (5) in the test cases to work with >> slow environments. Due to the failure on patchwork. >> >> - Remove systemd related commands in the test case. >> >> Major changes from the previous patchset [2]: >> >> - Split helpers. >> >> - fib6_set_expires() -> fib6_set_expires() and fib6_add_gc_list(). >> >> - fib6_clean_expires() -> fib6_clean_expires() and >> fib6_remove_gc_list(). >> >> - Fix rt6_add_dflt_router() to avoid racing of setting expires. >> >> - Remove unnecessary calling to fib6_clean_expires() in >> ip6_route_info_create(). >> >> - Add test cases of toggling routes between permanent and temporary >> and handling routes from RA messages. >> >> - Clean up routes by deleting the existing device and adding a new >> one. >> >> - Fix a potential issue in modify_prefix_route(). > > Note that we have a selftest failure in the batch including this series > for the fib_tests: > > https://netdev-3.bots.linux.dev/vmksft-net/results/456022/6-fib-tests-sh/stdout > > I haven't digged much, but I fear its related. Please have a look. > > For more info on how to reproduce the selftest environment: > > https://github.com/linux-netdev/nipa/wiki/How-to-run-netdev-selftests-CI-style > > Thanks, > > Paolo > I will check it.
On 2/8/24 03:55, Paolo Abeni wrote: > Note that we have a selftest failure in the batch including this series > for the fib_tests: > > https://netdev-3.bots.linux.dev/vmksft-net/results/456022/6-fib-tests-sh/stdout > > I haven't digged much, but I fear its related. Please have a look. > > For more info on how to reproduce the selftest environment: > > https://github.com/linux-netdev/nipa/wiki/How-to-run-netdev-selftests-CI-style > > Thanks, > > Paolo > To David and Hangbin, This specific case failed for the first time and passed on all following rounds. It is very time sensitive. Do you think it is OK to run "sysctl -wq net.ipv6.route.flush=1" to force a synchronized gc?
From: Kui-Feng Lee <thinker.li@gmail.com> This patchset is resent due to previous reverting. [1] FIB6 GC walks trees of fib6_tables to remove expired routes. Walking a tree can be expensive if the number of routes in a table is big, even if most of them are permanent. Checking routes in a separated list of routes having expiration will avoid this potential issue. Background ========== The size of a Linux IPv6 routing table can become a big problem if not managed appropriately. Now, Linux has a garbage collector to remove expired routes periodically. However, this may lead to a situation in which the routing path is blocked for a long period due to an excessive number of routes. For example, years ago, there is a commit c7bb4b89033b ("ipv6: tcp: drop silly ICMPv6 packet too big messages"). The root cause is that malicious ICMPv6 packets were sent back for every small packet sent to them. These packets add routes with an expiration time that prompts the GC to periodically check all routes in the tables, including permanent ones. Why Route Expires ================= Users can add IPv6 routes with an expiration time manually. However, the Neighbor Discovery protocol may also generate routes that can expire. For example, Router Advertisement (RA) messages may create a default route with an expiration time. [RFC 4861] For IPv4, it is not possible to set an expiration time for a route, and there is no RA, so there is no need to worry about such issues. Create Routes with Expires ========================== You can create routes with expires with the command. For example, ip -6 route add 2001:b000:591::3 via fe80::5054:ff:fe12:3457 \ dev enp0s3 expires 30 The route that has been generated will be deleted automatically in 30 seconds. GC of FIB6 ========== The function called fib6_run_gc() is responsible for performing garbage collection (GC) for the Linux IPv6 stack. It checks for the expiration of every route by traversing the trees of routing tables. The time taken to traverse a routing table increases with its size. Holding the routing table lock during traversal is particularly undesirable. Therefore, it is preferable to keep the lock for the shortest possible duration. Solution ======== The cause of the issue is keeping the routing table locked during the traversal of large trees. To solve this problem, we can create a separate list of routes that have expiration. This will prevent GC from checking permanent routes. Result ====== We conducted a test to measure the execution times of fib6_gc_timer_cb() and observed that it enhances the GC of FIB6. During the test, we added permanent routes with the following numbers: 1000, 3000, 6000, and 9000. Additionally, we added a route with an expiration time. Here are the average execution times for the kernel without the patch. - 120020 ns with 1000 permanent routes - 308920 ns with 3000 ... - 581470 ns with 6000 ... - 855310 ns with 9000 ... The kernel with the patch consistently takes around 14000 ns to execute, regardless of the number of permanent routes that are installed. Major changes from v4: - Fix the comment of fib6_add_gc_list(). Major changes from v3: - Move the checks of f6i->fib6_node to fib6_add_gc_list(). - Make spin_lock_bh() and spin_unlock_bh() stands out. - Explain the reason of the changes in the commit message of the patch 4. Major changes from v2: - Refactory the boilerplate checks in the test case. - check_rt_num() and check_rt_num_clean() Major changes from v1: - Reduce the numbers of routes (5) in the test cases to work with slow environments. Due to the failure on patchwork. - Remove systemd related commands in the test case. Major changes from the previous patchset [2]: - Split helpers. - fib6_set_expires() -> fib6_set_expires() and fib6_add_gc_list(). - fib6_clean_expires() -> fib6_clean_expires() and fib6_remove_gc_list(). - Fix rt6_add_dflt_router() to avoid racing of setting expires. - Remove unnecessary calling to fib6_clean_expires() in ip6_route_info_create(). - Add test cases of toggling routes between permanent and temporary and handling routes from RA messages. - Clean up routes by deleting the existing device and adding a new one. - Fix a potential issue in modify_prefix_route(). --- [1] https://lore.kernel.org/all/20231219030243.25687-1-dsahern@kernel.org/ [2] https://lore.kernel.org/all/20230815180706.772638-1-thinker.li@gmail.com/ v1: https://lore.kernel.org/all/20240131064041.3445212-1-thinker.li@gmail.com/ v2: https://lore.kernel.org/all/20240201082024.1018011-1-thinker.li@gmail.com/ v3: https://lore.kernel.org/all/20240202082200.227031-1-thinker.li@gmail.com/ v4: https://lore.kernel.org/all/20240205214033.937814-1-thinker.li@gmail.com/ Kui-Feng Lee (5): net/ipv6: set expires in rt6_add_dflt_router(). net/ipv6: Remove unnecessary clean. net/ipv6: Remove expired routes with a separated list of routes. net/ipv6: set expires in modify_prefix_route() if RTF_EXPIRES is set. selftests/net: Adding test cases of replacing routes and route advertisements. include/net/ip6_fib.h | 46 ++++++- include/net/ip6_route.h | 3 +- net/ipv6/addrconf.c | 41 +++++-- net/ipv6/ip6_fib.c | 60 ++++++++- net/ipv6/ndisc.c | 13 +- net/ipv6/route.c | 19 ++- tools/testing/selftests/net/fib_tests.sh | 148 +++++++++++++++++++---- 7 files changed, 287 insertions(+), 43 deletions(-)