nihilist/tor
1
0
Fork 0
mirror of https://gitlab.torproject.org/tpo/core/tor.git synced 2024-11-30 23:53:32 +01:00
Code Issues Packages Projects Releases Wiki Activity

add a few items it would be smart todo

Browse Source 
svn:r6457
This commit is contained in:
Roger Dingledine 2006-05-22 04:44:57 +00:00
parent 33e92cd5fb
commit 08403e105d
1 changed files with 22 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
doc/TODO
View File

@ -50,21 +50,28 @@ R - track down the patch for cross-compiling.
- Interim things: - Interim things:
- provide no-cache no-index headers from the dirport? - provide no-cache no-index headers from the dirport?
o remove down/useless descriptors from v1 directory?
- exitlist should avoid outputting the same IP address twice. - exitlist should avoid outputting the same IP address twice.
- if we have no predicted ports, don't fetch router descriptors.
this way we are more dormant.
Must-have items for 0.1.2.x: Must-have items for 0.1.2.x:
- If we fail to connect via an exit enclave, (warn and) try again
without demanding that exit node.
- If we have no predicted ports, don't fetch router descriptors.
This way we are more dormant.
- Directory guards - Directory guards
R - Server usability R - Server usability
- look into "uncounting" bytes spent on local connections. so
we can bandwidthrate but still have fast downloads.
- Write limiting; separate token bucket for write
- dir answers include a your-ip-address-is header, so we can
break our dependency on dyndns.
N - Better hidden service performance, with possible redesign. N - Better hidden service performance, with possible redesign.
- Asynchronous DNS - Asynchronous DNS
- What to use? C-ares? Libdns? AGL's patch? - What to use? C-ares? Libdns? AGL's patch?
- Better estimates in the directory of whether servers have good uptime - Better estimates in the directory of whether servers have good uptime
(high expected time to failure) or good guard qualities (high (high expected time to failure) or good guard qualities (high
fractional uptime). fractional uptime).
- AKA Track uptime as %-of-time-up, as well as time-since-last-down.
N . memory usage on dir servers. copy less! N . memory usage on dir servers. copy less!
o Remember offset and location of each descriptor in the cache/journal o Remember offset and location of each descriptor in the cache/journal
- When sending a big pile of descs to a client, don't shove them all on - When sending a big pile of descs to a client, don't shove them all on
@ -73,7 +80,7 @@ N . memory usage on dir servers. copy less!
refcounted. (Only slightly; we'd only need to know whether it's on refcounted. (Only slightly; we'd only need to know whether it's on
the routerlist->routers or routerlist->old_routers, and how many the routerlist->routers or routerlist->old_routers, and how many
directory conns are returning it.) directory conns are returning it.)
- Possibly, we could just to keep a list of the descriptor digests for - Possibly, we could just keep a list of the descriptor digests for
the descriptors we still want to send. We might end up truncating the descriptors we still want to send. We might end up truncating
some replies by returning fewer descriptors than were requested (if some replies by returning fewer descriptors than were requested (if
somebody requests a desc that we throw away before we deliver it), somebody requests a desc that we throw away before we deliver it),
@ -88,6 +95,8 @@ N . memory usage on dir servers. copy less!
N - oprofile including kernel time on multiple platforms N - oprofile including kernel time on multiple platforms
Items for 0.1.2: Items for 0.1.2:
- We need a separate list of "hidserv authorities" if we want to
retire moria1 from the main list.
- Improve controller - Improve controller
- change circuit status events to give more details, like purpose, - change circuit status events to give more details, like purpose,
whether they're internal, when they become dirty, when they become whether they're internal, when they become dirty, when they become
@ -114,7 +123,7 @@ Items for 0.1.2:
without using SOCKS. without using SOCKS.
- Make everything work with hidden services - Make everything work with hidden services
- Clients should refer to avoid exit nodes for non-exit path positions. - Clients should prefer to avoid exit nodes for non-exit path positions.
(bug 200) (bug 200)
- Make "setconf" and "hup" behavior cleaner for LINELIST config - Make "setconf" and "hup" behavior cleaner for LINELIST config
options (e.g. Log). Bug 238. options (e.g. Log). Bug 238.
@ -125,10 +134,7 @@ Items for 0.1.2:
- Design - Design
- Implement - Implement
- Have a "ReallyFast" status flag that means it. - Have a "Faster" status flag that means it. Fast2, Fast4, Fast8?
R - look into "uncounting" bytes spent on local connections. so
we can bandwidthrate but still have fast downloads.
- When we connect to a Tor server, it sends back a cell listing - When we connect to a Tor server, it sends back a cell listing
the IP it believes it is using. Use this to block dvorak's attack. the IP it believes it is using. Use this to block dvorak's attack.
@ -158,10 +164,10 @@ Topics to think about during 0.1.2.x development:
- Figure out hidden services. - Figure out hidden services.
Minor items for 0.1.2.x as time permits. Minor items for 0.1.2.x as time permits.
- Streamline how we define a guard node as 'up'. document it - The bw_accounting file should get merged into the state file.
somewhere. - Streamline how we define a guard node as 'up'. document it somewhere.
- Better installers and build processes. - Better installers and build processes.
- Commit edmanm's win32 makefil eto tor cvs contrib, or write a new one. - Commit edmanm's win32 makefile to tor cvs contrib, or write a new one.
- Can we cross-compile? - Can we cross-compile?
R - Christian Grothoff's attack of infinite-length circuit. R - Christian Grothoff's attack of infinite-length circuit.
the solution is to have a separate 'extend-data' cell type the solution is to have a separate 'extend-data' cell type
@ -177,7 +183,8 @@ N - Display the reasons in 'destroy' and 'truncated' cells under some
- We need a getrlimit equivalent on Windows so we can reserve some - We need a getrlimit equivalent on Windows so we can reserve some
file descriptors for saving files, etc. Otherwise we'll trigger file descriptors for saving files, etc. Otherwise we'll trigger
asserts when we're out of file descriptors and crash. asserts when we're out of file descriptors and crash.
- the tor client can do the "automatic proxy config url" thing? X the tor client can do the "automatic proxy config url" thing?
(no, let's leave this for applications like torbutton)
- Automatically determine what ports are reachable and start using - Automatically determine what ports are reachable and start using
those, if circuits aren't working and it's a pattern we recognize those, if circuits aren't working and it's a pattern we recognize
("port 443 worked once and port 9001 keeps not working"). ("port 443 worked once and port 9001 keeps not working").
@ -220,7 +227,7 @@ N - Vet all pending installer patches
- Clients should estimate their skew as median of skew from servers - Clients should estimate their skew as median of skew from servers
over last N seconds. over last N seconds.
- Security - Security
- Alices avoid duplicate class C nodes. - Alices avoid duplicate /24 servers.
- Analyze how bad the partitioning is or isn't. - Analyze how bad the partitioning is or isn't.
. Update the hidden service stuff for the new dir approach. . Update the hidden service stuff for the new dir approach.
@ -245,7 +252,6 @@ N - Vet all pending installer patches
- Make router_is_general_exit() a bit smarter once we're sure what it's for. - Make router_is_general_exit() a bit smarter once we're sure what it's for.
- rewrite how libevent does select() on win32 so it's not so very slow. - rewrite how libevent does select() on win32 so it's not so very slow.
- Write limiting; separate token bucket for write
- Audit everything to make sure rend and intro points are just as likely to - Audit everything to make sure rend and intro points are just as likely to
be us as not. be us as not.
- Do something to prevent spurious EXTEND cells from making middleman - Do something to prevent spurious EXTEND cells from making middleman
@ -261,7 +267,6 @@ Future version:
- DoS protection: TLS puzzles, public key ops, bandwidth exhaustion. - DoS protection: TLS puzzles, public key ops, bandwidth exhaustion.
- Specify? - Specify?
- tor-resolve script should use socks5 to get better error messages. - tor-resolve script should use socks5 to get better error messages.
- Track uptime as %-of-time-up, as well as time-since-last-down.
- hidserv offerers shouldn't need to define a SocksPort - hidserv offerers shouldn't need to define a SocksPort
* figure out what breaks for this, and do it. * figure out what breaks for this, and do it.
- Relax clique assumptions. - Relax clique assumptions.
@ -289,3 +294,4 @@ Blue-sky:
streams, at least according to the protocol. But we handle all that streams, at least according to the protocol. But we handle all that
we've seen in the wild. we've seen in the wild.
(Pending a user who needs this) (Pending a user who needs this)