Lab 3: Part A

Details

• An application calls Make(peers,me) to create a Paxos peer.
  • The peers argument contains the ports of all the peers (including this one),
  • The me argument is the index of this peer in the peers array.
• Start(seq,v) asks Paxos to start agreement on instance seq, with proposed value v;
  • Start() should return immediately, without waiting for agreement to complete.
  • The application calls Status(seq) to find out whether the Paxos peer thinks the instance has reached agreement,
    • and if so what the agreed value is.
• Status() should consult the local Paxos peer's state and return immediately;
  • it should not communicate with other peers.
  • the application may call Status() for old instances (but see the discussion of Done() below).
• implementation should be able to make progress on agreement for multiple instances at the same time.
  • if application peers call Start() with different sequence numbers at about the same time, your implementation should run the Paxos protocol concurrently for all of them.
  • you should not wait for agreement to complete for instance i before starting the protocol for instance i+1. Each instance should have its own separate execution of the Paxos protocol.
• long-running Paxos-based server must forget about instances that are no longer needed,
  • free the memory storing information about those instances
  • an instance is needed if the application still wants to be able to call Status() for it
    • or if another Paxos peer may not yet have reached agreement on that instance
  • when a particular peer application will no longer need to call Status() for any instance <= x, it should call Done(x).
    • that Paxos peer can't yet discard the instances, b/c. some other Paxos peer might not yet have agreed to the instance.
    • so each Paxos peer should tell each other peer the highest Done() argument supplied by its local application.
    • each Paxos peer will then have a Done() value from each other peer.
    • it should find the minimum, and discard all instances with sequence numbers <= that minimum.
    • The Min() method returns this minimum sequence number plus one.
    • it's OK for your Paxos to piggyback the Done() value in the agreement protocol packets;
    • that is, it's OK for peer P1 to only learn P2's latest Done() value the next time that P2 sends an agreement message to P1.
    • If Start() is called with a sequence number less than Min(), the Start() call should be ignored.
    • If Status() is called with a sequence number less than Min(), Status() should return Forgotten.

Plan

Here's a reasonable plan of attack:

1. Add elements to the Paxos struct in paxos.go to hold the state you'll need, according to the lecture pseudo-code.
   • You'll need to define a struct to hold information about each agreement instance.
2. Define RPC argument/reply type(s) for Paxos protocol messages, based on the lecture pseudo-code.
   • The RPCs must include the sequence number for the agreement instance to which they refer.
   • Remember the field names in the RPC structures must start with capital letters.
3. Write a proposer function that drives the Paxos protocol for an instance, and RPC handlers that implement acceptors.
   • Start a proposer function in its own thread for each instance, as needed (e.g. in Start()).
4. At this point you should be able to pass the first few tests.
5. Now implement forgetting.

Hints

Done: more than one Paxos instance may be executing at a given time, and they may be Start()ed and/or decided out of order (e.g. seq 10 may be decided before seq 5).

• before acting on log entry #i, wait for log entries < i to be decided

Done: in order to pass tests assuming unreliable network, your paxos should call the local acceptor through a function call rather than RPC.

Hint: remember that multiple application peers may call Start() on the same instance, perhaps with different proposed values. An application may even call Start() for an instance that has already been decided (maybe because it could race when issuing a NoOp for a hole in the log).

Hint: think about how your paxos will forget (discard) information about old instances before you start writing code. Each Paxos peer will need to store instance information in some data structure that allows individual instance records to be deleted (so that the Go garbage collector can free / re-use the memory).

Hint: you do not need to write code to handle the situation where a Paxos peer needs to re-start after a crash. If one of your Paxos peers crashes, it will never be re-started.

Hint: have each Paxos peer start a thread per un-decided instance whose job is to eventually drive the instance to agreement, by acting as a proposer.

Hint: a single Paxos peer may be acting simultaneously as acceptor and proposer for the same instance. Keep these two activities as separate as possible.

Hint: a proposer needs a way to choose a higher proposal number than any seen so far. This is a reasonable exception to the rule that proposer and acceptor should be separate. It may also be useful for the propose RPC handler to return the highest known proposal number if it rejects an RPC, to help the caller pick a higher one next time. The px.me value will be different in each Paxos peer, so you can use px.me to help ensure that proposal numbers are unique.

Hint: figure out the minimum number of messages Paxos should use when reaching agreement in non-failure cases and make your implementation use that minimum.

Hint: the tester calls Kill() when it wants your Paxos to shut down; Kill() sets px.dead. You should check px.dead in any loops you have that might run for a while, and break out of the loop if px.dead is true. It's particularly important to do this in any long-running threads you create.

Lab 3: Part B

Notes

• kvpaxos replicas should stay identical;
  • the only exception is that some replicas may lag others if they are not reachable.
• if replica isn't reachable for a while, but then starts being reachable => should eventually catch up (learn about operations that it missed)
• kvpaxos client code should try different replicas it knows about until one responds
• kvpaxos replica that is part of a majority which can reach each other should be able to serve client requests
• provide sequential consistency to applications that use its client interface.
  • completed application calls to the Clerk.Get(), Clerk.Put(), and Clerk.Append() methods in kvpaxos/client.go must appear to have affected all replicas in the same order and have at-most-once semantics
  • A Clerk.Get() should see the value written by the most recent Clerk.Put() or Clerk.Append() (in that order) to the same key.
  • One consequence of this is that you must ensure that each application call to Clerk.Put() or Clerk.Append() must appear in that order just once (i.e., write the key/value database just once), even though internally your client.go may have to send RPCs multiple times until it finds a kvpaxos server replica that replies.

Plan

Here's a reasonable plan:

1. Fill in the Op struct in server.go with the "value" information that kvpaxos will use Paxos to agree on, for each client request.
   • Op field names must start with capital letters.
   • You should use Op structs as the agreed-on values
     • for example, you should pass Op structs to Paxos.Start().
     • Go's RPC can marshall/unmarshall Op structs
     • the call to gob.Register() in StartServer() teaches it how.
2. Implement the PutAppend() handler in server.go.
   • it should enter a Put or Append Op in the Paxos log (i.e., use Paxos to allocate a Paxos instance, whose value includes the key and value (so that other kvpaxoses know about the Put() or Append())).
   • An Append Paxos log entry should contain the Append's arguments, but not the resulting value, since the result might be large.
3. Implement a Get() handler.
   • It should enter a Get Op in the Paxos log, and then "interpret" the log before that point to make sure its key/value database reflects all recent Put()s.
4. Add code to cope with duplicate client requests
   • including situations where:
     • the client sends a request to one kvpaxos replica,
     • client times out waiting for a reply,
     • client re-sends the request to a different replica.
   • the client request should execute just once.
   • make sure that your scheme for duplicate detection frees server memory quickly
     • for example, by having the client tell the servers which RPCs it has heard a reply for
     • it's OK to piggyback this information on the next client request

Hints

Hint: your server should try to assign the next available Paxos instance (sequence number) to each incoming client RPC. However, some other kvpaxos replica may also be trying to use that instance for a different client's operation. So the kvpaxos server has to be prepared to try different instances.

Hint: your kvpaxos servers should not directly communicate; they should only interact with each other through the Paxos log.

Hint: as in Lab 2, you will need to uniquely identify client operations to ensure that they execute just once. Also as in Lab 2, you can assume that each clerk has only one outstanding Put,Get, or Append.

Hint: a kvpaxos server should not complete a Get() RPC if it is not part of a majority (so that it does not serve stale data). This means that each Get() (as well as each Put() and Append()) must involve Paxos agreement.

Hint: don't forget to call the Paxos Done() method when a kvpaxos has processed an instance and will no longer need it or any previous instance.

Hint: your code will need to wait for Paxos instances to complete agreement. The only way to do this is to periodically call Status(), sleeping between calls. How long to sleep? A good plan is to check quickly at first, and then more slowly:

  to := 10 * time.Millisecond
  for {
    status, _ := kv.px.Status(seq)
    if status == paxos.Decided {
      ...
      return 
    }
    time.Sleep(to)
    if to < 10 * time.Second {
      to *= 2
    }
  }

Hint: if one of your kvpaxos servers falls behind (i.e. did not participate in the agreement for some instance), it will later need to find out what (if anything) was agreed to. A reasonable way to to this is to call Start(), which will either discover the previously agreed-to value, or cause agreement to happen. Think about what value would be reasonable to pass to Start() in this situation.

• So we would need a NoOp for this, In case the fallen-behind node who wanted to discover previously agreed-to value actually succeeds in proposing one

[DONE] Hint: When the test fails, check for gob error (e.g. "rpc: writing response: gob: type not registered for interface ...") in the log because go doesn't consider the error fatal, although it is fatal for the lab.

Algorithm

Each KVP server has a copy of the DB and of the log

Note: We will not reimplement a log, because we should not have to! The log is already implicitly implemented in the paxos library from Part A. And a server will know how much of the log it can apply to its local DB by calling Min() on its Paxos peer.

Q: What's the high-level algorithm
A: A sketch:

• Client sends a request to one of the Paxos servers
• Each server has a nextSeq number that it thinks the next request should get
  • it's possible that servers are out of date on this number due to partitioning, delays, etc.
    • S1 and S2 both start with seq #0
    • S1 gets request from C1
    • S2 gets request from C2
    • S1 and S2 will both propose the same seq. number
• Server S creates an Op op from the request and calls Paxos.Start(nextSeq, op) to get other peers to agree on this op
  • problem is they might not agree because they are also trying to get their own Op to be agreed upon
• If all (majority) agree, that's fine and dandy
  • TODO: Note that once a majority agrees there could still be a minority that needs to find out. How will they do it?
    • The proposer S who got the majority to agree will not finish his Start() loop until everyone receives the Decided message, so it seems that we do not need to do anything?
    • But what if that proposer S is down and there is still a majority up that can inform the minority about the agreement?
      • seems like the minority would need to ask about the status
• If they do not agree with S's value (i.e. S's proposal failed), then S needs to retry with a different nextSeq number.
  • at this point S can definitely increment his nextSeq number because he knows a value was agreed on
  • what should be nextSeq number that S picks if he fails proposing?
    • seems like it should just be the next one?

Q: Do we need to apply the log to the DB?
A: Yes, from the memory test cases and from the Piazza answers.

Q: When do we apply the log to the DB?
A: Seems like we can call Paxos.Min() and see which log entries everyone agreed on AND everyone HAS and apply those to our DB?

• however we have to be careful to keep the Paxos inst./seq. number the same, even after applying those entries just so we don't end up handling older Paxos agreements with the same seq. number

Q: Duplicate detection?
A: Only one outstanding client call => next client call can refer to previous call's XID and tell the server to get rid of it?