0.2.0 alpha #138

drunkirishcoder · 2021-03-29T22:58:32Z

Description

Finally here. Long waited divorce from Tokio pre-alpha as the underlying runtime. Along the way we also wanted to re-envision how we process packets. Major breaking changes,

goodbye Tokio (closing Upgrade the task scheduler to stable Tokio #75), hello Smol. this actually doesn't impact the end user much.
brand new runtime with a simpler set of API! all FFI code has been rewritten from the ground up too.
no more combinators. that was an idea carried over from NetBricks and that concept no longer align with how we handle types.
same example apps but they really work! no more replaying pcap files, they now interact with live traffic in our vagrant.

a couple things still missing

metrics! need to put metrics back.
revisit the packet ownership around parse and peek.

Type of change

Breaking change
Documentation update

codecov · 2021-03-29T23:08:31Z

Codecov Report

Merging #138 (1232a4c) into master (63b7155) will increase coverage by 7.60%.
The diff coverage is 55.59%.

@@            Coverage Diff             @@
##           master     #138      +/-   ##
==========================================
+ Coverage   68.69%   76.29%   +7.60%     
==========================================
  Files          66       50      -16     
  Lines        5836     5050     -786     
==========================================
- Hits         4009     3853     -156     
+ Misses       1827     1197     -630

Impacted Files	Coverage Δ
core/src/packets/arp.rs	`63.31% <ø> (-0.37%)`	⬇️
core/src/packets/ethernet.rs	`79.86% <0.00%> (ø)`
core/src/packets/icmp/v4/echo_reply.rs	`70.66% <ø> (ø)`
core/src/packets/icmp/v4/echo_request.rs	`80.00% <ø> (ø)`
core/src/packets/icmp/v4/mod.rs	`80.73% <ø> (ø)`
core/src/packets/icmp/v4/redirect.rs	`75.82% <ø> (ø)`
core/src/packets/icmp/v4/time_exceeded.rs	`74.07% <ø> (ø)`
core/src/packets/icmp/v6/echo_reply.rs	`70.66% <ø> (ø)`
core/src/packets/icmp/v6/echo_request.rs	`70.66% <ø> (ø)`
core/src/packets/icmp/v6/mod.rs	`77.96% <ø> (ø)`
... and 39 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

JakkuSakura · 2021-04-10T05:02:10Z

I have operation not supported running any of the examples.

drunkirishcoder · 2021-04-10T15:37:54Z

are you running the examples in our vagrant environment? they only work inside vagrant+docker setup because they have to assume a certain network topology with specific device names. if you are using vagrant, can you post the program output? I will take a look.

JakkuSakura · 2021-04-10T15:43:49Z

I'm running examples in the same environment where I run 0.1 examples. Old version runs just fine. note that I modified the device name and device id according to my network. I can give you access to my testing machine if you like

…

------------------ Original ------------------ From: Daniel Jin ***@***.***> Date: Sat, Apr 10, 2021 11:38 PM To: capsule-rs/capsule ***@***.***> Cc: QiuJiangkun ***@***.***>, Comment ***@***.***> Subject: Re: [capsule-rs/capsule] 0.2.0 alpha (#138) are you running the examples in our vagrant environment? they only work inside vagrant/docker setup because they have to assume a certain network topology with specific device names. if you are using vagrant, can you post the program output? I will take a look. — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

drunkirishcoder · 2021-04-10T15:50:31Z

At this time, we are only going to support the 0.2.0 examples running in our controlled environment. To make them work and interact with live packets, we have to make a lot of assumptions about their device name, IP address, MAC address, subnet etc. basically a lot goes into controlling how the packets are routed to make the examples work.

but if you set the log level to DEBUG and share the output, I can take a look and maybe identify what "operation not supported" is referring to. that error could suggest you are potentially encountering some device capability limitation? I'm not sure. but I don't have a lot info to go on.

JakkuSakura · 2021-04-10T16:08:09Z

Maybe my environment was not clean enough. After rebooting, all examples are runnable.

zeeshanlakhani

Still going through stuff and testing on baremetal, but starting through the process.

bench/combinators.rs

zeeshanlakhani · 2021-04-12T13:59:48Z

core/src/ffi/mod.rs

+    }
+}
+
+// delete later?


@drunkirishcoder just not something we're using currently?

hmmm honestly don't remember why I commented this, been almost a year. lol. I guess I had to do something during migrating code from v0.1 to v0.2? I will see what happens if I delete this.

removed comment. it is being used. might clean up in the following work around mbuf clones.

core/src/testils/mod.rs

core/src/runtime/pcap_dump.rs

zeeshanlakhani · 2022-01-10T11:03:55Z

core/src/ffi/mod.rs

+
+/// Makes NonNull even easier to use with the downside that it hides the
+/// unsafeness of the underlying pointer access.
+pub(crate) struct EasyPtr<T>(NonNull<T>);


Is there a better alternative to investigate here?

probably not? this is completely internal to ~~capsule~~ ffi. it just cuts down on how many unsafe {} we have to sprinkle through out the code base and the use is completely limited to the ffi layer. for example, all the packet headers are held as NonNull, not EasyPtr. if you see any usage outside the ffi layer, call it out.

Will double check

core/src/ffi/pcap.rs

core/src/packets/size_of.rs

Cargo.toml

zeeshanlakhani · 2022-02-16T11:14:43Z

core/Cargo.toml

 libc = "0.2"
-metrics-core = { version = "0.5", optional = true }
-metrics-runtime = { version = "0.13", optional = true, default-features = false }
+metrics = "0.14"


@drunkirishcoder should we remove for now (until we return metrics separately)?

zeeshanlakhani

so a few minor updates to make, but think we're ready to bring this in.

Will finish running this on the baremetal linux box as well.

zeeshanlakhani · 2022-02-16T11:17:32Z

@drunkirishcoder also maybe do a deps update and see if we can fix the security issue related to chrono/time.

drunkirishcoder · 2022-03-05T14:45:03Z

sigh

getting blocked by an indirect dependency. not sure which lib is pulling it in, but will need to update the sandbox to the latest stable rust. will do it when I find time soon.

zeeshanlakhani self-requested a review March 29, 2021 23:24

drunkirishcoder marked this pull request as draft March 29, 2021 23:25

drunkirishcoder linked an issue Mar 30, 2021 that may be closed by this pull request

Upgrade the task scheduler to stable Tokio #75

Open

drunkirishcoder marked this pull request as ready for review April 7, 2021 15:46

zeeshanlakhani reviewed Jan 10, 2022

View reviewed changes

zeeshanlakhani self-requested a review January 10, 2022 11:16

zeeshanlakhani reviewed Feb 16, 2022

View reviewed changes

zeeshanlakhani approved these changes Feb 16, 2022

View reviewed changes

drunkirishcoder force-pushed the djin/0.2.0-alpha branch 2 times, most recently from 8e818aa to 74ab2b5 Compare February 26, 2022 03:48

drunkirishcoder force-pushed the djin/0.2.0-alpha branch 2 times, most recently from a8a0686 to e76ccb7 Compare July 15, 2022 19:47

drunkirishcoder force-pushed the djin/0.2.0-alpha branch from e76ccb7 to beb33cb Compare September 27, 2022 18:48

drunkirishcoder added 8 commits September 27, 2022 15:43

new pattern for ffi

0355170

new core runtime

836e686

new mbuf allocation

71688c7

new port rx and tx loops

6288f0e

completely revamp examples

df52a1a

redo packet capture dump

565b611

general cleanup

87ef83e

ignore sanitize error

1232a4c

drunkirishcoder force-pushed the djin/0.2.0-alpha branch from bffcbb9 to 1232a4c Compare September 27, 2022 19:44

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

0.2.0 alpha #138

0.2.0 alpha #138

drunkirishcoder commented Mar 29, 2021 •

edited by zeeshanlakhani

Loading

codecov bot commented Mar 29, 2021 •

edited

Loading

JakkuSakura commented Apr 10, 2021

drunkirishcoder commented Apr 10, 2021 •

edited

Loading

JakkuSakura commented Apr 10, 2021 via email

drunkirishcoder commented Apr 10, 2021 •

edited

Loading

JakkuSakura commented Apr 10, 2021

zeeshanlakhani left a comment

zeeshanlakhani Apr 12, 2021

drunkirishcoder Jan 10, 2022

drunkirishcoder Mar 5, 2022

zeeshanlakhani Jan 10, 2022

drunkirishcoder Jan 10, 2022 •

edited

Loading

zeeshanlakhani Jan 10, 2022

zeeshanlakhani Feb 16, 2022

zeeshanlakhani left a comment

zeeshanlakhani commented Feb 16, 2022

drunkirishcoder commented Mar 5, 2022

0.2.0 alpha #138

Are you sure you want to change the base?

0.2.0 alpha #138

Conversation

drunkirishcoder commented Mar 29, 2021 • edited by zeeshanlakhani Loading

Description

Type of change

codecov bot commented Mar 29, 2021 • edited Loading

Codecov Report

JakkuSakura commented Apr 10, 2021

drunkirishcoder commented Apr 10, 2021 • edited Loading

JakkuSakura commented Apr 10, 2021 via email

drunkirishcoder commented Apr 10, 2021 • edited Loading

JakkuSakura commented Apr 10, 2021

zeeshanlakhani left a comment

Choose a reason for hiding this comment

zeeshanlakhani Apr 12, 2021

Choose a reason for hiding this comment

drunkirishcoder Jan 10, 2022

Choose a reason for hiding this comment

drunkirishcoder Mar 5, 2022

Choose a reason for hiding this comment

zeeshanlakhani Jan 10, 2022

Choose a reason for hiding this comment

drunkirishcoder Jan 10, 2022 • edited Loading

Choose a reason for hiding this comment

zeeshanlakhani Jan 10, 2022

Choose a reason for hiding this comment

zeeshanlakhani Feb 16, 2022

Choose a reason for hiding this comment

zeeshanlakhani left a comment

Choose a reason for hiding this comment

zeeshanlakhani commented Feb 16, 2022

drunkirishcoder commented Mar 5, 2022

drunkirishcoder commented Mar 29, 2021 •

edited by zeeshanlakhani

Loading

codecov bot commented Mar 29, 2021 •

edited

Loading

drunkirishcoder commented Apr 10, 2021 •

edited

Loading

drunkirishcoder commented Apr 10, 2021 •

edited

Loading

drunkirishcoder Jan 10, 2022 •

edited

Loading