|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [Xen-devel] [OSSTEST PATCH 5/8] Schema: Support database schema updates
See schema/README.schema, introduced in this patch, for the design.
Signed-off-by: Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx>
---
v2: Slight increase schema update name length format.
Docs fixes:
Change erroneous `three' to `four'.
Change `state' to `status' throghout.
Explain scope of <status>.
Sort out (and renumber) `Update order for Populate-then-rely'.
Sort out "Statuses" explanations.
Encourage use of DML update, rather than ad-hoc scripts,
for populating new columns.
---
Osstest/Executive.pm | 82 +++++++++++++++
mg-schema-create | 20 ++++
mg-schema-test-database | 6 +-
mg-schema-update | 255 +++++++++++++++++++++++++++++++++++++++++++++
schema/README.updates | 179 +++++++++++++++++++++++++++++++
schema/schema-updates.sql | 6 ++
6 files changed, 547 insertions(+), 1 deletion(-)
create mode 100755 mg-schema-update
create mode 100644 schema/README.updates
create mode 100644 schema/schema-updates.sql
diff --git a/Osstest/Executive.pm b/Osstest/Executive.pm
index fcef83f..e1fbe3b 100644
--- a/Osstest/Executive.pm
+++ b/Osstest/Executive.pm
@@ -56,6 +56,7 @@ BEGIN {
resource_check_allocated resource_shared_mark_ready
duration_estimator
db_pg_dsn opendb opendb_state
+ db_schema_updates_applied db_schema_updates_intree
);
%EXPORT_TAGS = ( colours => [qw($green $red $yellow $purple $blue)] );
@@ -128,6 +129,87 @@ sub grabrepolock_reexec {
}
}
+sub db_schema_updates_applied (;$) {
+ my ($cond) = @_;
+ my $r;
+ $cond //= '1=1';
+ eval {
+ local $dbh_tests->{PrintError} = 0;
+ $r = $dbh_tests->selectall_arrayref(<<END);
+ SELECT updatename, applytime
+ FROM schema_updates WHERE $cond
+END
+ };
+ if ($@) {
+ die unless
+ $dbh_tests->err()==7 && # DBD::Pg(3pm)
+ $dbh_tests->state() eq '42P01';
+ # http://www.postgresql.org/docs/current/static/errcodes-appendix.html
+ $r = [ ];
+ }
+ my @r;
+ foreach (@$r) {
+ push @r, { Name => $_->[0], Applied => $_->[1] };
+ }
+ return \@r;
+}
+
+sub db_schema_updates_intree (;$) {
+ my ($incommit) = @_;
+ # ->[]{Name}
+ # ->[]{Seq}
+ # ->[]{State}
+
+ my @results;
+
+ my @files;
+ if (!$incommit) {
+ @files = <schema/*.sql>;
+ } else {
+ local $/ = "\0";
+ open GLF, "-|", qw(git ls-tree -z), $incommit, "schema/" or die $!;
+ while (<GLF>) {
+ chomp;
+ next unless s/^\d+ blob \w+\t//;
+ push @files, $_;
+ }
+ $!=0; $?=0; close GLF or die "$! $? ($incommit)";
+ }
+
+ FILE: foreach my $f (@files) {
+ $f =~ m/\.sql$/ or next;
+ $f =~ m#/([a-z][0-9a-z-]+)\.sql$# or die "badly named .sql file $f\n";
+ my $name = $1;
+ next if $name eq 'initial';
+ if ($incommit) {
+ open SQLF, "-|", qw(git cat-file blob), "$incommit:$f" or die $!;
+ } else {
+ open SQLF, "<", $f or die "$f $!";
+ }
+ while (<SQLF>) {
+ chomp;
+ my $origl = $_;
+ next unless s/^\s*--\s*##OSSTEST##\s+//;
+ m/^0*([1-9]\d*)\s+(Harmless|Preparatory|Unfinished|Ready|Needed)\b/
+ or die "$origl ?";
+ push @results, {
+ Name => $name,
+ Seq => $1+0,
+ State => $2,
+ };
+ next FILE;
+ }
+ $!=0; $?=0; close SQLF; die "$f \`$name' no token ($! $?)";
+ }
+
+ @results = sort {
+ $a->{Seq} <=> $b->{Seq} ||
+ die "$a->{Name} $a->{Seq} == $b->{Name} $b->{Seq}"
+ } @results;
+
+ return \@results;
+}
+
#---------- database access ----------#
sub opendb_state () {
diff --git a/mg-schema-create b/mg-schema-create
index 54f1c76..1ee007b 100755
--- a/mg-schema-create
+++ b/mg-schema-create
@@ -30,6 +30,9 @@
# Options:
#
# -q don't print progress messages
+# --no-updates apply no schema updates
+# --stop-before --stop-after only apply some schema updates -
+# see mg-schema-update
set -e
set -o posix
@@ -38,6 +41,8 @@ set -o pipefail
progress () { printf "%s\n" "$*"; }
progress=progress
quietopt=''
+do_updates=true
+updates=()
while [ $# != 0 ]; do
arg=$1; shift
@@ -46,16 +51,31 @@ while [ $# != 0 ]; do
progress=:
quietopt=-q
;;
+ --stop-before|--stop-after)
+ updates+=("$arg" "$1"); shift
+ ;;
+ --stop-before=*|--stop-after=*)
+ updates+=("$arg"); shift
+ ;;
+ --no-updates)
+ do_updates=false
+ ;;
*)
echo >&2 "bad usage ($arg)"; exit 127
;;
esac
done
+export OSSTEST_DB_USEREAL_IGNORETEST='.*'
+
. ./cri-getconfig
$progress "Populating database..."
$(get_psql_cmd) $quietopt -f schema/initial.sql
+if $do_updates; then
+ ./mg-schema-update $quietopt apply-all "${updates[@]}"
+fi
+
$progress "Database set up."
diff --git a/mg-schema-test-database b/mg-schema-test-database
index 3616c4d..5c6a935 100755
--- a/mg-schema-test-database
+++ b/mg-schema-test-database
@@ -344,6 +344,8 @@ END
# Keep a copy as it came from dump, for comparison
cp $t.schema $t.schema.orig
+ wantupdates=$(./mg-schema-update list-applied)
+
# http://www.postgresql.org/message-id/26790.1306355327@xxxxxxxxxxxxx
perl -i~ -pe '
s/^/--/ if
@@ -391,7 +393,9 @@ END
psql_do <<END
CREATE DATABASE $dbname;
END
- withtest ./mg-schema-create -q
+ withtest ./mg-schema-create -q --no-updates
+
+ withtest ./mg-schema-update -q apply $wantupdates
printf ".\n"
diff --git a/mg-schema-update b/mg-schema-update
new file mode 100755
index 0000000..2b472fc
--- /dev/null
+++ b/mg-schema-update
@@ -0,0 +1,255 @@
+#!/usr/bin/perl -w
+
+# This is part of "osstest", an automated testing framework for Xen.
+# Copyright (C) 2009-2015 Citrix Inc.
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+
+# Usages:
+#
+# ./mg-schema-update [<options>] apply [<updatename>...]
+# ./mg-schema-update [<options>] show
+# ./mg-schema-update [<options>] apply-all
+#
+# Usual rune for applying updates:
+#
+# ./mg-schema-update -o <oldest-running-git-ref-spec> apply-all
+#
+# Options:
+#
+# -o<git-ref-spec> --oldest=<git-ref-spec>
+# Specify the oldest version of osstest that is currently running
+# anywhere against this DB. Used to determine compatibility of
+# updates.
+#
+# -f
+# Force. May be repeated for greater effect:
+#
+# Force installation of updates which break some old code, even
+# though -o was not specified so it is not known whether such old
+# code is still running anywhere.
+#
+# -ff
+# Force installation of updates which are known to break running
+# code (including, perhaps, this very code right here).
+#
+# -fff
+# Force an attempt to install an already-installed update. This
+# is not likely to work.
+
+
+use strict qw(vars);
+use DBI;
+use Osstest;
+use Osstest::Executive;
+
+use Getopt::Long qw(:config bundling gnu_compat require_order no_ignore_case);
+
+csreadconfig();
+
+our (%state, @state);
+our $there;
+our $force=0;
+our $quiet=0;
+
+sub getstate () {
+ my $record = sub {
+ my ($key,$list) = @_;
+ foreach my $entry (@$list) {
+ $state{ $entry->{Name} }{ $key } = $entry;
+ }
+ };
+
+ $record->('Db', db_schema_updates_applied());
+ $record->('Here', db_schema_updates_intree());
+ $record->('There', db_schema_updates_intree($there)) if $there;
+
+ foreach my $name (keys %state) {
+ my $out = $state{$name};
+ $out->{Name} = $name;
+ $out->{Sortkey} =
+ $out->{Here}{Seq} //
+ $out->{There}{Seq} //
+ 0;
+ }
+
+ foreach my $out (values %state) {
+ my $st = $out->{Here}{State} // 'missing';
+ # ->{Todo} = 0: no; 1: maybe (?old code); 2: always
+ # -1: already applied
+ if ($out->{Db}{Applied}) {
+ $out->{Todo} = -1;
+ $out->{Msg} = "already applied";
+ } elsif ($st =~ m/Harmless|Preparatory/) {
+ $out->{Todo} = 2;
+ $out->{Msg} = "right away";
+ } elsif ($st =~ m/Ready|Needed/) {
+ my $tst = $out->{There}{State} // 'missing';
+ if (!$there) {
+ $out->{Todo} = 1;
+ $out->{Msg} = "would break any old code";
+ } elsif ($tst =~ m/Harmless|Preparatory/) {
+ $out->{Todo} = 2;
+ $out->{Msg} = "specified revision can cope";
+ } else {
+ $out->{Todo} = 0;
+ $out->{Msg} = "specified revision would break";
+ }
+ } else {
+ $out->{Todo} = 0;
+ $out->{Msg} = "not ready";
+ }
+ die unless defined $out->{Todo} && defined $out->{Msg};
+
+ $out->{File} = "schema/$out->{Name}.sql";
+ }
+
+ @state = sort { $a->{Sortkey} <=> $b->{Sortkey} } values %state;
+}
+
+sub cmd_list_applied () {
+ die if @ARGV;
+ getstate();
+ foreach my $v (@state) {
+ next unless $v->{Db}{Applied};
+ print $v->{Name}, "\n" or die $!;
+ }
+}
+
+sub cmd_show () {
+ die if @ARGV;
+
+ getstate();
+
+ printf "%-25s", "Name";
+ printf " %-9s", "Worktree";
+ printf " %-11.11s", (sprintf "%8.11s", $there) if $there;
+ printf " DB update\n";
+
+ foreach my $v (@state) {
+ printf(" %-25s %5s %-5.5s",
+ $v->{Name},
+ $v->{Here}{Seq} // '',
+ $v->{Here}{State} // '-');
+ printf(" %5s %-5.5s",
+ $v->{There}{Seq} // '',
+ $v->{There}{State} // '-') if $there;
+ my $app;
+ if ($v->{Db}{Applied}) {
+ $app = show_abs_time($v->{Db}{Applied});
+ } else {
+ $app = (qw(No: Maybe: Ready!))[$v->{Todo}];
+ $app .= " ";
+ $app .= $v->{Msg};
+ }
+ printf(" %s\n", $app);
+ }
+
+ STDOUT->error and die $!;
+}
+
+sub want_apply ($) {
+ my ($v) = @_;
+ $v->{Todo} >= 2-$force;
+}
+
+sub applyone ($) {
+ my ($v) = @_;
+ die "Will not apply $v->{Name}.sql: $v->{Msg}\n"
+ unless want_apply($v);
+
+ my $fn = $v->{File};
+
+ db_retry($dbh_tests, \@all_lock_tables, sub {
+ print "Applying $fn...\n" unless $quiet;
+ open F, "<", $fn or die "$fn: $!";
+ local $/ = undef;
+ my $sql = <F>;
+ F->error and die $!;
+ close F or die $!;
+
+ $dbh_tests->do(<<END) if $quiet;
+ SET client_min_messages = warning;
+END
+
+ $dbh_tests->do($sql);
+
+ $dbh_tests->do(<<END, {}, $v->{Name}, time);
+ INSERT INTO schema_updates
+ (updatename, applytime)
+ VALUES (?, ?)
+END
+ });
+
+ print "Applying $fn done.\n" unless $quiet;
+}
+
+sub cmd_apply () {
+ print "No updates applied by calling apply with no update names.\n"
+ unless $quiet;
+ foreach my $name (@ARGV) {
+ getstate();
+ my $ent = $state{$name};
+ die "unknown update \`$name'\n" unless $ent;
+ die "update \`$name' not in this tree\n" unless $ent->{Here}{Seq};
+ applyone($ent);
+ }
+}
+
+sub cmd_apply_all () {
+ my (%stopafter, %stopbefore);
+
+ GetOptions('--stop-after=s' => sub { $stopafter{$_[1]} = 1 },
+ '--stop-before=s' => sub { $stopbefore{$_[1]} = 1 });
+ die "further arguments to apply-all prohibited\n" if @ARGV;
+
+ getstate();
+ foreach my $v (@state) {
+ next unless $v->{Here}{Seq};
+ my $stop = sub {
+ my ($map) = @_;
+ return 0 unless
+ $map->{ $v->{Name} } ||
+ $map->{ $v->{Here}{Seq} };
+ print "Stopping at $v->{Name} ($v->{Here}{Seq}).\n"
+ unless $quiet;
+ return 1;
+ };
+
+ last if $stop->(\%stopbefore);
+
+ if (want_apply($v)) {
+ applyone($v);
+ } else {
+ print "Skipping $v->{File}: $v->{Msg}\n" unless $quiet;
+ }
+
+ last if $stop->(\%stopafter);
+ }
+
+ print "Appropriate updates applied.\n" unless $quiet;
+}
+
+GetOptions('f|force+' => \$force,
+ 'q+' => \$quiet,
+ 'o|oldest=s' => \$there);
+
+die "need operation\n" unless @ARGV;
+
+my $subcmd= shift @ARGV;
+$subcmd =~ s/-/_/g;
+my $subcmdproc = ${*::}{"cmd_$subcmd"};
+die "unknown subcommand" unless $subcmdproc;
+$subcmdproc->();
diff --git a/schema/README.updates b/schema/README.updates
new file mode 100644
index 0000000..622410c
--- /dev/null
+++ b/schema/README.updates
@@ -0,0 +1,179 @@
+SCHEMA DEFINITION AND SCHEMA UPDATES (PRODUCTION `EXECUTIVE' MODE)
+==================================================================
+
+To generate a new DB, we apply the original schema (in initial.sql)
+and then apply all the updates, in order.
+
+We maintain a table in the DB which records which updates are applied.
+
+
+Schema update snippet format
+----------------------------
+
+Schema update snippets should be called
+ schema/<updatename>.sql
+
+They should contain DDL commands (ALTER TABLE etc.) to make whatever
+changes are needed.
+
+They MUST NOT contain BEGIN or COMMIT.
+
+They must contain a special comment near the top:
+
+ -- ##OSSTEST## <sequence> <status>
+
+<updatename> is a string (/^[a-z][0-9a-z-]+$/) which uniquely identifies
+the update. It must not be changed because existing installations
+rely on updates having stable names.
+
+<sequence> is a positive integer, which should be unique. Updates are
+applied in order.
+
+<status> reflects the compatibility of various schema versions. It is
+a literal string naming one of the statuses shown in `Update orders',
+below.
+
+<status> depends on the nature of the specific database change, and
+the behaviour and capabilities of the other code in the same revision
+of osstest.git. But, so <status> does not depend on the state of the
+database. Applying a schema update to a database does not change its
+`status'.
+
+In principle, each update can separately be in applied or not applied
+in any one moment in time (in various databases), and simultaneously
+have different statuses in different relevant versions of osstest.git.
+So overall the possible states of an update in the whole world are the
+cross product of (i) status in each relevant osstest revision, and
+(ii) appliedness (boolean) in each relevant database instance.
+
+
+Update orders
+-------------
+
+There are four reasonable plans for schema changes:
+
+ * Fully intercompatible: both old code and new code are each
+ compatible with both old schema and new schema. The code and
+ schema updates may be done in any order.
+
+ Such a schema change always has status:
+ Harmless
+
+ * Explicit conditional: first update the code to understand both
+ versions of the schema; then update the schema; then drop the
+ compatibility code.
+
+ Such a schema change always has status:
+ Unfinished (or absent) in old code
+ Ready in intermediate code
+ Needed in the final code
+
+ * Code first: the new code works with either old or new schema,
+ but the old code cannot cope with the new schema.
+
+ Such a schema change has status:
+ Unfinished (or absent) in old code
+ Ready in new code
+
+ * Schema first: the new schema works with any code; but the old
+ schema does not work with new code.
+
+ Such a schema change has status:
+ Preparatory in old code
+ Needed in the new code
+
+
+Update order for Populate-then-rely
+-----------------------------------
+
+This is for when we want to record new information and then later rely
+on it. There are typically two schema changes:
+
+* To add the column(s). I will call this `add'. It is a `Schema
+ first' change, in the taxonomy above.
+
+* To add appropriate constraints, to prevent the new information being
+ left blank. I will call this `constraint'. This is a `Code first'
+ or `Explicit conditional' change in the taxonomy above.
+
+1. Commit: new schema update `add', status Preparatory.
+
+2. Commit: new schema update `constraint', status Unfinished.
+
+3. Apply: `add'.
+
+4. Optionally commit: code to read new column, but which tolerates
+ both complete absence of the column, and/or it containing NULL
+ (or whatever the DEFAULT value is).
+
+5. Commit: code to populate new column; changing `add' to status
+ Needed and `constraint' to status Ready.
+
+6. Optionally commit: code which read new column, but which tolerates
+ it containing NULL/DEFAULT. (`add' is already Needed.)
+
+7. If necessary commit: idempotent utility script to populate missing
+ data. (Alternatively, this can be done with DML statements in the
+ `constraint' schema update .sql file. This is better if it is
+ possible.)
+
+8. Wait for all executions of old code to finish. (This obviously
+ implies first getting a push of all the commits mentioned above.)
+
+8. If necessary, execute utility script to populate missing data.
+
+9. Apply: `constraint'.
+
+10. Optionally commit: code which relies on new column, and does not
+ necessarily tolerate NULL/DEFAULT; changing `constraint' to Needed.
+
+
+`Commit' means committing somewhere public and probably pushing to
+osstest.git#pretest, but not necessarily getting a push. (It
+necessarily precedes any formal testing of the relevant changes on a
+production instance.)
+
+`Apply' (and `execute utility script') should only be done using a
+properly acked version of osstest.git. If verifying the sanity of the
+schema change is nontrivial then ad-hoc tests may need to have been
+run with a testing instance of the database. Using only a pushed
+production version is a good idea to avoid the possibility that the
+production database might contain changes which are not evident in
+published code (or worse, which are different in future versions).
+
+Subject to those conditions, `Apply' means an administrator running
+./mg-schema-update as osstest; if `wait for executions of old code to
+finish is needed', this will usually involve passing an appropriate
+`-o' option.
+
+
+Statuses and rules for push and db update
+-----------------------------------------
+
+ Harmless
+ Preparatory
+ No restrictions
+
+ Unfinished
+ (sql fragment entirely missing is equivalent to Unfinished)
+ Schema update: prevented
+ Code push: unrestricted
+
+ Ready
+ Schema update: need all live code to be Preparatory/Ready/Needed
+ Code push: unrestricted
+
+ Needed
+ Schema update: need all live code to be Preparatory/Ready/Needed
+ Code push: depends on schema update
+
+
+"Code push: depends on schema update" is not currently implemented.
+However, many (most?) such changes would cause the push gate itself to
+fail.
+
+"Need all live code to be ..." means to look for the status of this
+schema update in other running versions of osstest. An attempt at
+this is provided in the form of the `-o' option to mg-schema-update.
+It is the administrator's responsibility to select an appropriate
+argument to `-o'.
diff --git a/schema/schema-updates.sql b/schema/schema-updates.sql
new file mode 100644
index 0000000..cd8dc0c
--- /dev/null
+++ b/schema/schema-updates.sql
@@ -0,0 +1,6 @@
+-- ##OSSTEST## 001 Harmless
+
+CREATE TABLE schema_updates (
+ updatename TEXT PRIMARY KEY,
+ applytime integer NOT NULL
+);
--
1.7.10.4
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel
|
 |
Lists.xenproject.org is hosted with RackSpace, monitoring our |