libdb/docs/programmer_reference/rep_trans.html
2011-09-13 13:44:24 -04:00

328 lines
22 KiB
HTML
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<title>Transactional guarantees</title>
<link rel="stylesheet" href="gettingStarted.css" type="text/css" />
<meta name="generator" content="DocBook XSL Stylesheets V1.73.2" />
<link rel="start" href="index.html" title="Berkeley DB Programmer's Reference Guide" />
<link rel="up" href="rep.html" title="Chapter 12.  Berkeley DB Replication" />
<link rel="prev" href="rep_bulk.html" title="Bulk transfer" />
<link rel="next" href="rep_lease.html" title="Master Leases" />
</head>
<body>
<div xmlns="" class="navheader">
<div class="libver">
<p>Library Version 11.2.5.2</p>
</div>
<table width="100%" summary="Navigation header">
<tr>
<th colspan="3" align="center">Transactional guarantees</th>
</tr>
<tr>
<td width="20%" align="left"><a accesskey="p" href="rep_bulk.html">Prev</a> </td>
<th width="60%" align="center">Chapter 12. 
Berkeley DB Replication
</th>
<td width="20%" align="right"> <a accesskey="n" href="rep_lease.html">Next</a></td>
</tr>
</table>
<hr />
</div>
<div class="sect1" lang="en" xml:lang="en">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a id="rep_trans"></a>Transactional guarantees</h2>
</div>
</div>
</div>
<p>It is important to consider replication in the context of the overall
database environment's transactional guarantees. To briefly review,
transactional guarantees in a non-replicated application are based on
the writing of log file records to "stable storage", usually a disk
drive. If the application or system then fails, the Berkeley DB logging
information is reviewed during recovery, and the databases are updated
so that all changes made as part of committed transactions appear, and
all changes made as part of uncommitted transactions do not appear. In
this case, no information will have been lost.</p>
<p>If a database environment does not require the log be flushed to
stable storage on transaction commit (using the <a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a>
flag to increase performance at the cost of sacrificing transactional
durability), Berkeley DB recovery will only be able to restore the system to
the state of the last commit found on stable storage. In this case,
information may have been lost (for example, the changes made by some
committed transactions may not appear in the databases after recovery).</p>
<p>Further, if there is database or log file loss or corruption (for
example, if a disk drive fails), then catastrophic recovery is
necessary, and Berkeley DB recovery will only be able to restore the system
to the state of the last archived log file. In this case, information
may also have been lost.</p>
<p>Replicating the database environment extends this model, by adding a
new component to "stable storage": the client's replicated information.
If a database environment is replicated, there is no lost information
in the case of database or log file loss, because the replicated system
can be configured to contain a complete set of databases and log records
up to the point of failure. A database environment that loses a disk
drive can have the drive replaced, and it can then rejoin the
replication group.</p>
<p>Because of this new component of stable storage, specifying
<a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> in a replicated environment no longer sacrifices
durability, as long as one or more clients have acknowledged receipt of
the messages sent by the master. Since network connections are often
faster than local synchronous disk writes, replication becomes a way
for applications to significantly improve their performance as well as
their reliability.</p>
<p>The return status from the application's <span class="bold"><strong>send</strong></span> function must be
set by the application to ensure the transactional guarantees the
application wants to provide. Whenever the <span class="bold"><strong>send</strong></span> function
returns failure, the local database environment's log is flushed as
necessary to ensure that any information critical to database integrity
is not lost. Because this flush is an expensive operation in terms of
database performance, applications should avoid returning an error from
the <span class="bold"><strong>send</strong></span> function, if at all possible.</p>
<p>The only interesting message type for replication transactional
guarantees is when the application's <span class="bold"><strong>send</strong></span> function was called
with the <a href="../api_reference/C/reptransport.html#transport_DB_REP_PERMANENT" class="olink">DB_REP_PERMANENT</a> flag specified. There is no reason
for the <span class="bold"><strong>send</strong></span> function to ever return failure unless the
<a href="../api_reference/C/reptransport.html#transport_DB_REP_PERMANENT" class="olink">DB_REP_PERMANENT</a> flag was specified -- messages without the
<a href="../api_reference/C/reptransport.html#transport_DB_REP_PERMANENT" class="olink">DB_REP_PERMANENT</a> flag do not make visible changes to databases,
and the <span class="bold"><strong>send</strong></span> function can return success to Berkeley DB as soon as
the message has been sent to the client(s) or even just copied to local
application memory in preparation for being sent.</p>
<p>When a client receives a <a href="../api_reference/C/reptransport.html#transport_DB_REP_PERMANENT" class="olink">DB_REP_PERMANENT</a> message, the client
will flush its log to stable storage before returning (unless the client
environment has been configured with the <a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> option).
If the client is unable to flush a complete transactional record to disk
for any reason (for example, there is a missing log record before the
flagged message), the call to the <a href="../api_reference/C/repmessage.html" class="olink">DB_ENV-&gt;rep_process_message()</a> method on the client
will return <a href="../api_reference/C/repmessage.html#repmsg_DB_REP_NOTPERM" class="olink">DB_REP_NOTPERM</a> and return the LSN of this record
to the application in the <span class="bold"><strong>ret_lsnp</strong></span> parameter.
The application's client or master
message handling loops should take proper action to ensure the correct
transactional guarantees in this case. When missing records arrive
and allow subsequent processing of previously stored permanent
records, the call to the <a href="../api_reference/C/repmessage.html" class="olink">DB_ENV-&gt;rep_process_message()</a> method on the client will
return <a href="../api_reference/C/repmessage.html#repmsg_DB_REP_ISPERM" class="olink">DB_REP_ISPERM</a> and return the largest LSN of the
permanent records that were flushed to disk. Client applications
can use these LSNs to know definitively if any particular LSN is
permanently stored or not.</p>
<p>An application relying on a client's ability to become a master and
guarantee that no data has been lost will need to write the <span class="bold"><strong>send</strong></span>
function to return an error whenever it cannot guarantee the site that
will win the next election has the record. Applications not requiring
this level of transactional guarantees need not have the <span class="bold"><strong>send</strong></span>
function return failure (unless the master's database environment has
been configured with <a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a>), as any information critical
to database integrity has already been flushed to the local log before
<span class="bold"><strong>send</strong></span> was called.</p>
<p>To sum up, the only reason for the <span class="bold"><strong>send</strong></span> function to return
failure is when the master database environment has been configured to
not synchronously flush the log on transaction commit (that is,
<a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> was configured on the master), the
<a href="../api_reference/C/reptransport.html#transport_DB_REP_PERMANENT" class="olink">DB_REP_PERMANENT</a> flag is specified for the message, and the
<span class="bold"><strong>send</strong></span> function was unable to determine that some number of
clients have received the current message (and all messages preceding
the current message). How many clients need to receive the message
before the <span class="bold"><strong>send</strong></span> function can return success is an application
choice (and may not depend as much on a specific number of clients
reporting success as one or more geographically distributed clients).</p>
<p>If, however, the application does require on-disk durability on the master,
the master should be configured to synchronously flush the log on commit.
If clients are not configured to synchronously flush the log,
that is, if a client is running with <a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> configured,
then it is up to the application to reconfigure that client
appropriately when it becomes a master. That is, the
application must explicitly call <a href="../api_reference/C/envset_flags.html" class="olink">DB_ENV-&gt;set_flags()</a> to
disable asynchronous log flushing as part of re-configuring
the client as the new master.</p>
<p>Of course, it is important to ensure that the replicated master and
client environments are truly independent of each other. For example,
it does not help matters that a client has acknowledged receipt of a
message if both master and clients are on the same power supply, as the
failure of the power supply will still potentially lose information.</p>
<p>Configuring your replication-based application to achieve the proper
mix of performance and transactional guarantees can be complex. In
brief, there are a few controls an application can set to configure the
guarantees it makes: specification of <a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> for the
master environment, specification of <a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> for the
client environment, the priorities of different sites participating in
an election, and the behavior of the application's <span class="bold"><strong>send</strong></span>
function.</p>
<p>Applications using Replication Manager are free to use
<a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> at the master and/or clients as they see fit. The
behavior of the <span class="bold"><strong>send</strong></span> function that Replication Manager provides
on the application's behalf is determined by an "acknowledgement
policy", which is configured by the <a href="../api_reference/C/repmgrset_ack_policy.html" class="olink">DB_ENV-&gt;repmgr_set_ack_policy()</a> method.
Clients always send acknowledgements for <a href="../api_reference/C/reptransport.html#transport_DB_REP_PERMANENT" class="olink">DB_REP_PERMANENT</a>
messages (unless the acknowledgement policy in effect indicates that the
master doesn't care about them). For a <a href="../api_reference/C/reptransport.html#transport_DB_REP_PERMANENT" class="olink">DB_REP_PERMANENT</a>
message, the master blocks the sending thread until either it receives
the proper number of acknowledgements, or the <a href="../api_reference/C/repset_timeout.html#set_timeout_DB_REP_ACK_TIMEOUT" class="olink">DB_REP_ACK_TIMEOUT</a>
expires. In the case of timeout, Replication Manager returns an error
code from the <span class="bold"><strong>send</strong></span> function, causing Berkeley DB to flush the
transaction log before returning to the application, as previously
described. The default acknowledgement policy is
<a href="../api_reference/C/repmgrset_ack_policy.html#ackspolicy_DB_REPMGR_ACKS_QUORUM" class="olink">DB_REPMGR_ACKS_QUORUM</a>, which ensures that the effect of a
permanent record remains durable following an election.</p>
<p>First, it is rarely useful to write and synchronously flush the log when
a transaction commits on a replication client. It may be useful where
systems share resources and multiple systems commonly fail at the same
time. By default, all Berkeley DB database environments, whether master or
client, synchronously flush the log on transaction commit or prepare.
Generally, replication masters and clients turn log flush off for
transaction commit using the <a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> flag.</p>
<p>Consider two systems connected by a network interface. One acts as the
master, the other as a read-only client. The client takes over as
master if the master crashes and the master rejoins the replication
group after such a failure. Both master and client are configured to
not synchronously flush the log on transaction commit (that is,
<a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> was configured on both systems). The
application's <span class="bold"><strong>send</strong></span> function never returns failure to the Berkeley DB
library, simply forwarding messages to the client (perhaps over a
broadcast mechanism), and always returning success. On the client, any
<a href="../api_reference/C/repmessage.html#repmsg_DB_REP_NOTPERM" class="olink">DB_REP_NOTPERM</a> returns from the client's <a href="../api_reference/C/repmessage.html" class="olink">DB_ENV-&gt;rep_process_message()</a> method
are ignored, as well. This system configuration has excellent
performance, but may lose data in some failure modes.</p>
<p>If both the master and the client crash at once, it is possible to lose
committed transactions, that is, transactional durability is not being
maintained. Reliability can be increased by providing separate power
supplies for the systems and placing them in separate physical locations.</p>
<p>If the connection between the two machines fails (or just some number
of messages are lost), and subsequently the master crashes, it is
possible to lose committed transactions. Again, transactional
durability is not being maintained. Reliability can be improved in a
couple of ways:</p>
<div class="orderedlist">
<ol type="1">
<li>
<p>
Use a reliable network protocol (for example, TCP/IP instead of UDP).
</p>
</li>
<li>
<p>
Increase the number of clients and network paths to make it
less likely that a message will be lost. In this case, it is
important to also make sure a client that did receive the
message wins any subsequent election. If a client that did not
receive the message wins a subsequent election, data can still
be lost.
</p>
</li>
</ol>
</div>
<p>Further, systems may want to guarantee message delivery to the client(s)
(for example, to prevent a network connection from simply discarding
messages). Some systems may want to ensure clients never return
out-of-date information, that is, once a transaction commit returns
success on the master, no client will return old information to a
read-only query. Some of the following changes to a Base API application
may be used to address these issues:</p>
<div class="orderedlist">
<ol type="1">
<li>
<p>
Write the application's <span class="bold"><strong>send</strong></span>
function to not return to Berkeley DB until one or more clients
have acknowledged receipt of the message. The number of
clients chosen will be dependent on the application: you will
want to consider likely network partitions (ensure that a
client at each physical site receives the message) and
geographical diversity (ensure that a client on each coast
receives the message).
</p>
</li>
<li>
<p>
Write the client's message processing loop to not acknowledge
receipt of the message until a call to the <a href="../api_reference/C/repmessage.html" class="olink">DB_ENV-&gt;rep_process_message()</a> method
has returned success. Messages resulting in a return of
<a href="../api_reference/C/repmessage.html#repmsg_DB_REP_NOTPERM" class="olink">DB_REP_NOTPERM</a> from the <a href="../api_reference/C/repmessage.html" class="olink">DB_ENV-&gt;rep_process_message()</a> method mean the message
could not be flushed to the client's disk. If the client does
not acknowledge receipt of such messages to the master until a
subsequent call to the <a href="../api_reference/C/repmessage.html" class="olink">DB_ENV-&gt;rep_process_message()</a> method returns
<a href="../api_reference/C/repmessage.html#repmsg_DB_REP_ISPERM" class="olink">DB_REP_ISPERM</a> and the LSN returned is at least as large as
this message's LSN, then the master's <span class="bold"><strong>send</strong></span> function will not return
success to the Berkeley DB library. This means the thread
committing the transaction on the master will not be allowed to
proceed based on the transaction having committed until the
selected set of clients have received the message and consider
it complete.
</p>
<p>
Alternatively, the client's message processing loop could
acknowledge the message to the master, but with an error code
indicating that the application's <span class="bold"><strong>send</strong></span> function should not return to
the Berkeley DB library until a subsequent acknowledgement from
the same client indicates success.
</p>
<p>
The application send callback function invoked by Berkeley DB
contains an LSN of the record being sent (if appropriate for
that record). When <a href="../api_reference/C/repmessage.html" class="olink">DB_ENV-&gt;rep_process_message()</a> method returns indicators that
a permanent record has been written then it also returns the
maximum LSN of the permanent record written.
</p>
</li>
</ol>
</div>
<p>There is one final pair of failure scenarios to consider. First, it is
not possible to abort transactions after the application's <span class="bold"><strong>send</strong></span>
function has been called, as the master may have already written the
commit log records to disk, and so abort is no longer an option.
Second, a related problem is that even though the master will attempt
to flush the local log if the <span class="bold"><strong>send</strong></span> function returns failure,
that flush may fail (for example, when the local disk is full). Again,
the transaction cannot be aborted as one or more clients may have
committed the transaction even if <span class="bold"><strong>send</strong></span> returns failure. Rare
applications may not be able to tolerate these unlikely failure modes.
In that case the application may want to:</p>
<div class="orderedlist">
<ol type="1">
<li>
<p>
Configure the master to do always local synchronous commits
(turning off the <a href="../api_reference/C/envset_flags.html#envset_flags_DB_TXN_NOSYNC" class="olink">DB_TXN_NOSYNC</a> configuration). This will
decrease performance significantly, of course (one of the
reasons to use replication is to avoid local disk writes.) In
this configuration, failure to write the local log will cause
the transaction to abort in all cases.
</p>
</li>
<li>
<p>
Do not return from the application's <span class="bold"><strong>send</strong></span> function under any conditions,
until the selected set of clients has acknowledged the message.
Until the <span class="bold"><strong>send</strong></span> function
returns to the Berkeley DB library, the thread committing the
transaction on the master will wait, and so no application will
be able to act on the knowledge that the transaction has
committed.
</p>
</li>
</ol>
</div>
</div>
<div class="navfooter">
<hr />
<table width="100%" summary="Navigation footer">
<tr>
<td width="40%" align="left"><a accesskey="p" href="rep_bulk.html">Prev</a> </td>
<td width="20%" align="center">
<a accesskey="u" href="rep.html">Up</a>
</td>
<td width="40%" align="right"> <a accesskey="n" href="rep_lease.html">Next</a></td>
</tr>
<tr>
<td width="40%" align="left" valign="top">Bulk transfer </td>
<td width="20%" align="center">
<a accesskey="h" href="index.html">Home</a>
</td>
<td width="40%" align="right" valign="top"> Master Leases</td>
</tr>
</table>
</div>
</body>
</html>