cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Peter Schuller (Commented) (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-3829) make seeds *only* be seeds, not special in gossip
Date Fri, 03 Feb 2012 20:01:56 GMT


Peter Schuller commented on CASSANDRA-3829:

Historically, I believe this has been for ensuring partitions heal. However, with the ring
persisted after CASSANDRA-1518 this is probably not important in the common case of an established

If the premise is that you've created a partition you are kind of fubar:ed anyway (you will
have caused inconsistencies in data access). Even with "seeds as only be true seeds", you
should never be partitioned to begin with as long as the seeds, when you *do* need them (bootstrap)
is a correct list of members in the cluster.

Now, it is true that if you e.g. bring a node down, lots of changes happen to the ring ("everyone"
changes IP etc), and attempt to bring a node up again whose entire notion of the ring is no
longer valid (or same for a group of hosts), you will have a problem. Especially for small
clusters, this may be something to seriously consider.

There are, I think, reasonable mitigations here that I think will be needed anyway, having
to do with making node start-up include certain steps that ensure the node is in a valid ring
prior to taking client traffic - but I want to avoid getting into that discussion at the moment
(will revisit in the locator redesign JIRA:s).

This is probably not true, you can just change them everywhere and let the nodes restart naturally
due to whatever reason.

That's if you aren't blocking on it and don't care about loosing track of whether you're satisfying
the seed invariant. If you're wanting to maintain the "X seeds, no seed that is not actually
in the cluster" and you need to e.g. pop out one node that's a seed as part of e.g. ring re-adjustment
or just because you picked a few hosts (that were in a rack or whatever) to remove from the
ring, you now need to actually push the seed change out.

Again, I don't believe not doing this is a problem per say, but if we go with the officially
supported behavior/invariant of maintaining seeds.

I should of course mention that simply making the seeds dynamically re-load:able mitigates
this problem quite a lot, so if this is the main issue that is in fact an easy fix to make.
So this is not really my primary argument. I am much more interested in ensuring that we don't
rely on seeds for correct propagation through gossip, and removing the mental load on operators
to consider seeds for anything but initial bootstrapping of nodes.

I'm not sure I follow, why would a node told to autobootstrap disregard that and just join
the ring?

That's what Cassandra does - check {{joinTokenRing()}}. A node that detects that it is a seed
(i.e., it is itself listed in its own list of seeds) will skip the bootstrapping step and
just enter the ring. Very dangerous. For example, you loose a node due to a disk crash. It
comes back up with clean state. You deploy your confirmation. auto_bootstrap is true by default
(which is good and correct and let's not change that). You forget that it's one of the seeds.
*Bang* now you have brought a node into the cluster serving inconsistent reads.

> make seeds *only* be seeds, not special in gossip 
> --------------------------------------------------
>                 Key: CASSANDRA-3829
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Peter Schuller
>            Assignee: Peter Schuller
>            Priority: Minor
> First, a little bit of "framing" on how seeds work:
> The concept of "seed hosts" makes fundamental sense; you need to
> "seed" a new node with some information required in order to join a
> cluster. Seed hosts is the information Cassandra uses for this
> purpose.
> But seed hosts play a role even after the initial start-up of a new
> node in a ring. Specifically, seed hosts continue to be gossiped to
> separately by the Gossiper throughout the life of a node and the
> cluster.
> Generally, operators must be careful to ensure that all nodes in a
> cluster are appropriately configured to refer to an overlapping set of
> seed hosts. Strictly speaking this should not be necessary (see
> further down though), but is the general recommendation. An
> unfortunate side-effect of this is that whenever you are doing ring
> management, such as replacing nodes, removing nodes, etc, you have to
> keep in mind which nodes are seeds.
> For example, if you bring a new node into the cluster, doing
> everything right with token assignment and auto_bootstrap=true, it
> will just enter the cluster without bootstrap - causing inconsistent
> reads. This is dangerous.
> And worse - changing the notion of which nodes are seeds across a
> cluster requires a *rolling restart*. It can be argued that it should
> actually be okay for nodes other than the one being fiddled with to
> incorrectly treat the fiddled-with node as a seed node, but this fact
> is highly opaque to most users that are not intimately familiar with
> Cassandra internals.
> This adds additional complexity to operations, as it introduces a
> reason why you cannot view the ring as completely homogeneous, despite
> the fundamental idea of Cassandra that all nodes should be equal.
> Now, fast forward a bit to what we are doing over here to avoid this
> problem: We have a zookeeper based systems for keeping track of hosts
> in a cluster, which is used by our Cassandra client to discover nodes
> to talk to. This works well.
> In order to avoid the need to manually keep track of seeds, we wanted
> to make seeds be automatically discoverable in order to eliminate as
> an operational concern. We have implemented a seed provider that does
> this for us, based on the data we keep in zookeeper.
> We could see essentially three ways of plugging this in:
> * (1) We could simply rely on not needing overlapping seeds and grab whatever we have
when a node starts.
> * (2) We could do something like continually treat all other nodes as seeds by dynamically
changing the seed list (involves some other changes like having the Gossiper update it's notion
of seeds.
> * (3) We could completely eliminate the use of seeds *except* for the very specific purpose
of initial start-up of an unbootstrapped node, and keep using a static (for the duration of
the node's uptime) seed list.
> (3) was attractive because it felt like this was the original intent
> of seeds; that they be used for *seeding*, and not be constantly
> required during cluster operation once nodes are already joined.
> Now before I make the suggestion, let me explain how we are currently
> (though not yet in production) handling seeds and start-up.
> First, we have the following relevant cases to consider during a normal start-up:
> * (a) we are starting up a cluster for the very first time
> * (b) we are starting up a new clean node in order to join it to a pre-existing cluster
> * (c) we are starting up a pre-existing already joined node in a pre-existing cluster
> First, we proceeded on the assumption that we wanted to remove the use
> of seeds during regular gossip (other than on initial startup). This
> means that for the (c) case, we can *completely* ignore seeds. We
> never even have to discover the seed list, or if we do, we don't have
> to use them.
> This leaves (a) and (b). In both cases, the critical invariant we want
> to achieve is that we must have one or more *valid* seeds (valid means
> for (b) that the seed is in the cluster, and for (a) that it is one of
> the nodes that are part of the initial cluster setup).
> In the (c) case the problem is trivial - ignore seeds.
> In the (a) case, the algorithm is:
> * Register with zookeeper as a seed
> * Wait until we see *at least one* seed *other than ourselves* in zookeeper
> * Continue regular start-up process with the seed list (with 1 or more seeds)
> In the (b) case, the algorithm is:
> * Wait until we see *at least one* seed in zookeeper
> * Continue regular start-up process with the seed list (with 1 or more seeds)
> * Once fully up (around the time we listen to thrift), register as a seed in zookeeper
> With the annoyance that you have to explicitly let Cassandra know that
> "I am starting a cluster for the very first time from scratch", and
> ignoring the problem of single node clusters (just to avoid
> complicating this post further), this guarantees in both cases that
> all nodes eventually see each other.
> In the (a) case, all nodes except one are guaranteed to see the "one"
> node. The "one" node is guaranteed to see one of the others. Thus -
> convergence.
> In the (b) case, it's simple - the new node is guaranteed to see one
> or more nodes that are in the cluster - convergence.
> The current status is that we have implemented the seed provider and
> the start-up sequence works. But in order to simplify Cassandra (and
> to avoid having to diverge), we propose that we take this to its
> conclusion and officially make seeds only relevant on start-up, by
> only ever gossiping to seeds when in pre-bootstrap mode during
> start-up.
> The perceived benefits are:
> * Simplicity for the operator. All nodes are equal once joined; you can almost forget
completely about seeds.
> * No rolling restarts or potential for footshooting a node into a cluster without bootstrap
because it happened to be a seed.
> * Production clusters will suddenly start to actually *test* the gossip protocol without
relying on seeds. How sure are we that it even works, and that phi conviction is appropriate
and RING_DELAY is appropriate, given that practical clusters tend to gossip to a random (among
very few) seeds? This change would make it so that we *always* gossip randomly to anyone in
the cluster, and there should be no danger that a cluster happens to hold together because
seeds are up - only to explode when they are not.
> * It eliminates non-trivial concerns with automatic seed discover, particularly when
you want that seed discovery to be rack and DC aware. All you care about it what was described
above; if that seed happens to fail, we simply fail to find the cluster and can abort start-up
and it can be retried. There is no need for "redundancy" in seeds.
> Thoughts? Are seeds important (by design) in some way other than for seeding? What do
other people think about the implications of RING_DELAY etc?

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:!default.jspa
For more information on JIRA, see:


View raw message