In checkpointer and walwriter, avoid calling PostmasterIsAlive unless
WaitLatch has reported WL_POSTMASTER_DEATH. This saves a kernel call per
iteration of the process's outer loop, which is not all that much, but a
cycle shaved is a cycle earned. I had already removed the unconditional
PostmasterIsAlive calls in bgwriter and pgstat in previous patches, but
forgot that WL_POSTMASTER_DEATH is supposed to be treated as untrustworthy
(per comment in unix_latch.c); so adjust those two cases to match.
There are a few other places where the same idea might be applied, but only
after substantial code rearrangement, so I didn't bother.
Latch-ify the stats collector, so that it does not need an arbitrary wakeup
cycle to check for postmaster death. The incremental savings in idle power
is pretty marginal, since we only had it waking every two seconds; but I
believe that this patch may also improve the collector's performance under
load, by reducing the number of kernel calls made per message when messages
are arriving constantly (we now avoid a select/poll call except when we
need to sleep). The change also reduces the time needed for a normal
database shutdown on platforms where signals don't interrupt select().
This patch adjusts the core statistics views to match the decision already
taken for pg_stat_statements, that values representing elapsed time should
be represented as float8 and measured in milliseconds. By using float8,
we are no longer tied to a specific maximum precision of timing data.
(Internally, it's still microseconds, but we could now change that without
needing changes at the SQL level.)
The columns affected are
pg_stat_bgwriter.checkpoint_write_time
pg_stat_bgwriter.checkpoint_sync_time
pg_stat_database.blk_read_time
pg_stat_database.blk_write_time
pg_stat_user_functions.total_time
pg_stat_user_functions.self_time
pg_stat_xact_user_functions.total_time
pg_stat_xact_user_functions.self_time
The first four of these are new in 9.2, so there is no compatibility issue
from changing them. The others require a release note comment that they
are now double precision (and can show a fractional part) rather than
bigint as before; also their underlying statistics functions now match
the column definitions, instead of returning bigint microseconds.
Ants Aasma's original patch to add timing information for buffer I/O
requests exposed this data at the relation level, which was judged too
costly. I've here exposed it at the database level instead.
Add counters for number and size of temporary files used
for spill-to-disk queries for each database to the
pg_stat_database view.
Tomas Vondra, review by Magnus Hagander
This separates the state (running/idle/idleintransaction etc) into
it's own field ("state"), and leaves the query field containing just
query text.
The query text will now mean "current query" when a query is running
and "last query" in other states. Accordingly,the field has been
renamed from current_query to query.
Since backwards compatibility was broken anyway to make that, the procpid
field has also been renamed to pid - along with the same field in
pg_stat_replication for consistency.
Scott Mead and Magnus Hagander, review work from Greg Smith
This greatly reduces the WAL volume, especially when the table is narrow.
The overhead of locking the heap page is also reduced. Reduced WAL traffic
also makes it scale a lot better, if you run multiple COPY processes at
the same time.
Avoid possibly dumping core when pgstat_track_activity_query_size has a
less-than-default value; avoid uselessly searching for the query string
of a successfully-exited backend; don't bother putting out an ERRDETAIL if
we don't have a query to show; some other minor stylistic improvements.
To avoid minimize risk inside the postmaster, we subject this feature
to a number of significant limitations. We very much wish to avoid
doing any complex processing inside the postmaster, due to the
posssibility that the crashed backend has completely corrupted shared
memory. To that end, no encoding conversion is done; instead, we just
replace anything that doesn't look like an ASCII character with a
question mark. We limit the amount of data copied to 1024 characters,
and carefully sanity check the source of that data. While these
restrictions would doubtless be unacceptable in a general-purpose
logging facility, even this limited facility seems like an improvement
over the status quo ante.
Marti Raudsepp, reviewed by PDXPUG and myself
This is merely an exercise in satisfying pedants, not a bug fix, because
in every case we were checking for failure later with ferror(), or else
there was nothing useful to be done about a failure anyway. Document
the latter cases.
This reverts commit 79b2ee20c8, which proved
to not be very informative; it looks like the "pgstat wait timeout"
warnings in the buildfarm are just a symptom of running on heavily loaded
machines, and there isn't any weird mechanism causing them to appear.
To try to reduce the frequency of buildfarm failures from this effect,
increase PGSTAT_MAX_WAIT_TIME from 5 seconds to 10.
Also, arrange to not send a fresh inquiry message every single time through
the loop, as that seems more likely to cause problems (by swamping the
collector) than fix them. We'll now send an inquiry the first time through
the delay loop, and every 640 msec thereafter.
As per my recent proposal, this refactors things so that these typedefs and
macros are available in a header that can be included in frontend-ish code.
I also changed various headers that were undesirably including
utils/timestamp.h to include datatype/timestamp.h instead. Unsurprisingly,
this showed that half the system was getting utils/timestamp.h by way of
xlog.h.
No actual code changes here, just header refactoring.
This is in hopes of learning more about what causes "pgstat wait timeout"
warnings in the buildfarm. This patch should probably be reverted once
we've learned what we can. As coded, it will result in regression test
"failures" at half the delay that the existing code does, so I expect
to see a few more than before.
detect postmaster death. Postmaster keeps the write-end of the pipe open,
so when it dies, children get EOF in the read-end. That can conveniently
be waited for in select(), which allows eliminating some of the polling
loops that check for postmaster death. This patch doesn't yet change all
the loops to use the new mechanism, expect a follow-on patch to do that.
This changes the interface to WaitLatch, so that it takes as argument a
bitmask of events that it waits for. Possible events are latch set, timeout,
postmaster death, and socket becoming readable or writeable.
The pipe method behaves slightly differently from the kill() method
previously used in PostmasterIsAlive() in the case that postmaster has died,
but its parent has not yet read its exit code with waitpid(). The pipe
returns EOF as soon as the process dies, but kill() continues to return
true until waitpid() has been called (IOW while the process is a zombie).
Because of that, change PostmasterIsAlive() to use the pipe too, otherwise
WaitLatch() would return immediately with WL_POSTMASTER_DEATH, while
PostmasterIsAlive() would claim it's still alive. That could easily lead to
busy-waiting while postmaster is in zombie state.
Peter Geoghegan with further changes by me, reviewed by Fujii Masao and
Florian Pflug.
When we added the ability for vacuum to skip heap pages by consulting the
visibility map, we made it just not update the reltuples/relpages
statistics if it skipped any pages. But this could leave us with extremely
out-of-date stats for a table that contains any unchanging areas,
especially for TOAST tables which never get processed by ANALYZE. In
particular this could result in autovacuum making poor decisions about when
to process the table, as in recent report from Florian Helmberger. And in
general it's a bad idea to not update the stats at all. Instead, use the
previous values of reltuples/relpages as an estimate of the tuple density
in unvisited pages. This approach results in a "moving average" estimate
of reltuples, which should converge to the correct value over multiple
VACUUM and ANALYZE cycles even when individual measurements aren't very
good.
This new method for updating reltuples is used by both VACUUM and ANALYZE,
with the result that we no longer need the grotty interconnections that
caused ANALYZE to not update the stats depending on what had happened
in the parent VACUUM command.
Also, fix the logic for skipping all-visible pages during VACUUM so that it
looks ahead rather than behind to decide what to do, as per a suggestion
from Greg Stark. This eliminates useless scanning of all-visible pages at
the start of the relation or just after a not-all-visible page. In
particular, the first few pages of the relation will not be invariably
included in the scanned pages, which seems to help in not overweighting
them in the reltuples estimate.
Back-patch to 8.4, where the visibility map was introduced.
Tracks one counter for each database, which is reset whenever
the statistics for any individual object inside the database is
reset, and one counter for the background writer.
Tomas Vondra, reviewed by Greg Smith
We were failing to zero out some pg_stat_database counters that have
been added since the initial pgstats coding. This is a bug, but not
back-patching the fix since changing this behavior in a minor release
seems a cure worse than the disease.
Report and patch by Tomas Vondra.
This new field counts the number of times that a backend which writes a
buffer out to the OS must also fsync() it. This happens when the
bgwriter fsync request queue is full, and is generally detrimental to
performance, so it's good to know when it's happening. Along the way,
log a new message at level DEBUG1 whenever we fail to hand off an fsync,
so that the problem can also be seen in examination of log files
(if the logging level is cranked up high enough).
Greg Smith, with minor tweaks by me.
We may as well make pgstat_count_heap_scan() and related macros just count
whenever rel->pgstat_info isn't null. Testing pgstat_track_counts buys
nothing at all in the normal case where that flag is ON; and when it's OFF,
the pgstat_info link will be null, so it's still a useless test.
This change is unlikely to buy any noticeable performance improvement,
but a cycle shaved is a cycle earned; and my investigations earlier today
convinced me that we're down to the point where individual instructions in
the inner execution loops are starting to matter.
statistics counts. These numbers are being accumulated but haven't yet been
transmitted to the collector (and won't be, until the transaction ends).
For some purposes, though, it's handy to be able to look at them.
Joel Jacobson, reviewed by Itagaki Takahiro
unable to read a stats file for reasons other than ENOENT, and having to reset
last_statrequest because it's later than current time in the collector.
Not clear if this will shed any light on the "pgstat wait timeout" business,
but it seems like a good idea in general.
In passing, do some message-style-police work on recently-added
pgstat_reset_shared_counters code.
This silences some warnings on Win64. Not using the proper SOCKET datatype
was actually wrong on Win32 as well, but didn't cause any warnings there.
Also create define PGINVALID_SOCKET to indicate an invalid/non-existing
socket, instead of using a hardcoded -1 value.
decisions about when to auto-analyze.
The previous code depended on n_live_tuples + n_dead_tuples - last_anl_tuples,
where all three of these numbers could be bad estimates from ANALYZE itself.
Even worse, in the presence of a steady flow of HOT updates and matching
HOT-tuple reclamations, auto-analyze might never trigger at all, even if all
three numbers are exactly right, because n_dead_tuples could hold steady.
To fix, replace last_anl_tuples with an accurately tracked count of the total
number of committed tuple inserts + updates + deletes since the last ANALYZE
on the table. This can still be compared to the same threshold as before, but
it's much more trustworthy than the old computation. Tracking this requires
one more intra-transaction counter per modified table within backends, but no
additional memory space in the stats collector. There probably isn't any
measurable speed difference; if anything it might be a bit faster than before,
since I was able to eliminate some per-tuple arithmetic operations in favor of
adding sums once per (sub)transaction.
Also, simplify the logic around pgstat vacuum and analyze reporting messages
by not trying to fold VACUUM ANALYZE into a single pgstat message.
The original thought behind this patch was to allow scheduling of analyzes
on parent tables by artificially inflating their changes_since_analyze count.
I've left that for a separate patch since this change seems to stand on its
own merit.
The temporary hash tables made by pgstat_collect_oids should be allocated
in a short-term memory context, which is not the default behavior of
hash_create. Noted while looking through hash_create calls in connection
with Robert Haas' recent complaint.
This is a pre-existing bug, but it doesn't seem important enough to
back-patch. The hash table is not so large that it would matter unless this
happened many times within a session, which seems quite unlikely.
Formerly, these message types would be discarded unless there was already
a stats hash table entry for the target table. However, the intent of
saving hash table space for unused tables was subverted by the fact that
the physical I/O done by the vacuum or analyze would result in an immediately
following tabstat message, which would create the hash table entry anyway.
All that we had left was surprising loss of statistical data, as in a recent
complaint from Jaime Casanova.
It seems unlikely that a real database would have many tables that go totally
untouched over the long haul, so the consensus is that this "optimization"
serves little purpose anyhow. Remove it, and just create the hash table
entry on demand in all cases.
To make this work in the base case, pg_database now has a nailed-in-cache
relation descriptor that is initialized using hardwired knowledge in
relcache.c. This means pg_database is added to the set of relations that
need to have a Schema_pg_xxx macro maintained in pg_attribute.h. When this
path is taken, we'll have to do a seqscan of pg_database to find the row
we need.
In the normal case, we are able to do an indexscan to find the database's row
by name. This is made possible by storing a global relcache init file that
describes only the shared catalogs and their indexes (and therefore is usable
by all backends in any database). A new backend loads this cache file,
finds its database OID after an indexscan on pg_database, and then loads
the local relcache init file for that database.
This change should effectively eliminate number of databases as a factor
in backend startup time, even with large numbers of databases. However,
the real reason for doing it is as a first step towards getting rid of
the flat files altogether. There are still several other sub-projects
to be tackled before that can happen.
behavior in cases where we don't know the heap tuple count accurately; in
particular partial vacuum, but this also makes the API a bit more useful
for ANALYZE. This patch adds "estimated_count" flags to both structs so
that an approximate count can be flagged as such, and adjusts the logic
so that approximate counts are not used for updating pg_class.reltuples.
This fixes my previous complaint that VACUUM was putting ridiculous values
into pg_class.reltuples for indexes. The actual impact of that bug is
limited, because the planner only pays attention to reltuples for an index
if the index is partial; which probably explains why beta testers hadn't
noticed a degradation in plan quality from it. But it needs to be fixed.
The whole thing is a bit messy and should be redesigned in future, because
reltuples now has the potential to drift quite far away from reality when
a long period elapses with no non-partial vacuums. But this is as good as
it's going to get for 8.4.
skipped. We could update relpages anyway, but it seems better to only
update it together with reltuples, because we use the reltuples/relpages
ratio in the planner. Also don't update n_live_tuples in pgstat.
ANALYZE in VACUUM ANALYZE now needs to update pg_class, if the
VACUUM-phase didn't do so. Added some boolean-passing to let analyze_rel
know if it should update pg_class or not.
I also moved the relcache invalidation (to update rd_targblock) from
vac_update_relstats to where RelationTruncate is called, because
vac_update_relstats is not called for partial vacuums anymore. It's more
obvious to send the invalidation close to the truncation that requires it.
Per report by Ned T. Crigler.
where no function stats entries exist. Partial response to Pavel's
observation that small VACUUM operations are noticeably slower in CVS HEAD
than 8.3.
upon requests from backends, rather than on a fixed 500msec cycle. (There's
still throttling logic to ensure it writes no more often than once per
500msec, though.) This should result in a significant reduction in stats file
write traffic in typical scenarios where the stats are demanded only
infrequently.
This approach also means that the former difficulty with changing
stats_temp_directory on-the-fly has gone away, so remove the caution about
that as well as the thrashing we did to minimize the trouble window.
In passing, also fix pgstat_report_stat() so that we will send a stats
message if we have function call stats but not table stats to report;
this fixes a bug in the recent patch to support function-call stats.
Martin Pihlak
variable stats_temp_directory, instead of requiring the admin to
mount/symlink the pg_stat_tmp directory manually.
For now the config variable is PGC_POSTMASTER. Room for further improvment
that would allow it to be changed on-the-fly.
This allows the use of a ramdrive (either through mount or symlink) for
the temporary file that's written every half second, which should
reduce I/O.
On server shutdown/startup, the file is written to the old location in
the global directory, to preserve data across restarts.
Bump catversion since the $PGDATA directory layout changed.
As the buffer could now be a lot larger than before, and copying it could
thus be a lot more expensive than before, use strcpy instead of memcpy to
copy the query string, as was already suggested in comments. Also, only copy
the PgBackendStatus struct and string if the slot is in use.
Patch by Thomas Lee, with some changes by me.
corresponding struct definitions. This allows other headers to avoid including
certain highly-loaded headers such as rel.h and relscan.h, instead using just
relcache.h, heapam.h or genam.h, which are more lightweight and thus cause less
unnecessary dependencies.
functions.
Note that because this patch changes FmgrInfo, any external C functions
you might be testing with 8.4 will need to be recompiled.
Patch by Martin Pihlak, some editorialization by me (principally, removing
tracking of getrusage() numbers)
classed all as "dead"; also get it to count DEAD item pointers as dead rows,
instead of ignoring them as before. Also improve matters so that tuples
previously inserted or deleted by our own transaction are handled nicely:
the stats collector's live-tuple and dead-tuple counts will end up correct
after our transaction ends, regardless of whether we end in commit or abort.
While there's more work that could be done to improve the counting of in-doubt
tuples in both VACUUM and ANALYZE, this commit is enough to alleviate some
known bad behaviors in 8.3; and the other stuff that's been discussed seems
like research projects anyway.
Pavan Deolasee and Tom Lane
query texts only to the server log. This eliminates the issue of possible
leaking of security-sensitive data in other sessions' queries. Since the
log is presumed secure, we can now log the queries of all sessions involved
in the deadlock, whether or not they belong to the same user as the one
reporting the failure.
(if they'd be visible to the current user in pg_stat_activity).
This might look like it's subject to race conditions, but it's actually
pretty safe because at the time DeadLockReport() is constructing the
report, we haven't yet aborted our transaction and so we can expect that
everyone else involved in the deadlock is still blocked on some lock.
(There are corner cases where that might not be true, such as a statement
timeout triggering in another backend before we finish reporting; but at
worst we'd report a misleading activity string, so it seems acceptable
considering the usefulness of reporting the queries.)
Original patch by Itagaki Takahiro, heavily modified by me.
buffers that cannot possibly need to be cleaned, and estimates how many
buffers it should try to clean based on moving averages of recent allocation
requests and density of reusable buffers. The patch also adds a couple
more columns to pg_stat_bgwriter to help measure the effectiveness of the
bgwriter.
Greg Smith, building on his own work and ideas from several other people,
in particular a much older patch from Itagaki Takahiro.
* stats_start_collector goes away; we always start the collector process,
unless prevented by a problem with setting up the stats UDP socket.
* stats_reset_on_server_start goes away; it seems useless in view of the
availability of pg_stat_reset().
* stats_block_level and stats_row_level are merged into a single variable
"track_counts", which controls all reports sent to the collector process.
* stats_command_string is renamed to track_activities.
* log_autovacuum is renamed to log_autovacuum_min_duration to better reflect
its meaning.
The log_autovacuum change is not a compatibility issue since it didn't exist
before 8.3 anyway. The other changes need to be release-noted.
columns, and the new version can be stored on the same heap page, we no longer
generate extra index entries for the new version. Instead, index searches
follow the HOT-chain links to ensure they find the correct tuple version.
In addition, this patch introduces the ability to "prune" dead tuples on a
per-page basis, without having to do a complete VACUUM pass to recover space.
VACUUM is still needed to clean up dead index entries, however.
Pavan Deolasee, with help from a bunch of other people.
so that we will be able to create a cookie for all processes for CSVlogs.
It is set wherever MyProcPid is set. Take the opportunity to remove the now
unnecessary session-only restriction on the %s and %c escapes in log_line_prefix.
we don't know at that point which relation OID to tell pgstat to forget.
The code was passing the relfilenode, which is incorrect, and could possibly
cause some other relation's stats to be zeroed out. While we could try to
clean this up, it seems much simpler and more reliable to let the next
invocation of pgstat_vacuum_tabstat() fix things; which indeed is how it
worked before I introduced the buggy code into 8.1.3 and later :-(.
Problem noticed by Itagaki Takahiro, fix is per subsequent discussion.
over a fairly long period of time, rather than being spat out in a burst.
This happens only for background checkpoints carried out by the bgwriter;
other cases, such as a shutdown checkpoint, are still done at full speed.
Remove the "all buffers" scan in the bgwriter, and associated stats
infrastructure, since this seems no longer very useful when the checkpoint
itself is properly throttled.
Original patch by Itagaki Takahiro, reworked by Heikki Linnakangas,
and some minor API editorialization by me.
or abort within a backend; rearrange InitPostgres processing to make it so.
Revealed by just-added Asserts along with ECPG regression tests (hm, I wonder
why the core regression tests didn't expose it?). This possibly is another
reason for missing stats updates ...
and aborted transactions have different effects; also teach it not to assume
that prepared transactions are always committed.
Along the way, simplify the pgstats API by tying counting directly to
Relations; I cannot detect any redeeming social value in having stats
pointers in HeapScanDesc and IndexScanDesc structures. And fix a few
corner cases in which counts might be missed because the relation's
pgstat_info pointer hadn't been set.
messages to the stats collector. This avoids the problem that enabling
stats_row_level for autovacuum has a significant overhead for short
read-only transactions, as noted by Arjen van der Meijden. We can avoid
an extra gettimeofday call by piggybacking on the one done for WAL-logging
xact commit or abort (although that doesn't help read-only transactions,
since they don't WAL-log anything).
In my proposal for this, I noted that we could change the WAL log entries
for commit/abort to record full TimestampTz precision, instead of only
time_t as at present. That's not done in this patch, but will be committed
separately.
when a relation is opened multiple times in the same transaction. This is
particularly useful for system catalogs, which we may heap_open or index_open
many times in a transaction, and it doesn't really cost anything extra even
if the rel is touched but once. Motivated by study of an example from Greg
Stark, in which pgstat_initstats() accounted for an unreasonably large
fraction of the runtime.
its table list and then rechecks pgstat before vacuuming each table to
verify that no one has vacuumed the table in the meantime.
In the current autovacuum world this only means that a worker will not
vacuum a table that a user has vacuumed manually after the worker started.
When support for multiple autovacuum workers is introduced, this will reduce
the probability of simultaneous workers on the same database doing redundant
work.
continuously, and requests vacuum runs of "autovacuum workers" to postmaster.
The workers do the actual vacuum work. This allows for future improvements,
like allowing multiple autovacuum jobs running in parallel.
For now, the code keeps the original behavior of having a single autovac
process at any time by sleeping until the previous worker has finished.
entries for the victim database go away sooner rather than later. We already
did the equivalent thing at the per-relation level, not sure why it's not
been done for whole databases. With this change, pgstat_vacuum_tabstat
should usually not find anything to do; though we still need it as a backstop
in case DROPDB or TABPURGE messages get lost under load.
already collected in the current transaction; this allows plpgsql functions to
watch for stats updates even though they are confined to a single transaction.
Use this instead of the previous kluge involving pg_stat_file() to wait for
the stats collector to update in the stats regression test. Internally,
decouple storage of stats snapshots from transaction boundaries; they'll
now stick around until someone calls pgstat_clear_snapshot --- which xact.c
still does at transaction end, to maintain the previous behavior. This makes
the logic a lot cleaner, at the price of a couple dozen cycles per transaction
exit.
input in the stats collector. Our select() emulation is apparently buggy
for UDP sockets :-(. This should resolve problems with stats collection
(and hence autovacuum) failing under more than minimal load. Diagnosis
and patch by Magnus Hagander.
Patch probably needs to be back-ported to 8.1 and 8.0, but first let's
see if it makes the buildfarm happy...
(or other types of pg_class entry): the function pgstat_vacuum_tabstat,
invoked during VACUUM startup, had runtime proportional to the number of
stats table entries times the number of pg_class rows; in other words
O(N^2) if the stats collector's information is reasonably complete.
Replace list searching with a hash table to bring it back to O(N)
behavior. Per report from kim at myemma.com.
Back-patch as far as 8.1; 8.0 and before use different coding here.
identify long-running transactions. Since we already need to record
the transaction-start time (e.g. for now()), we don't need any
additional system calls to report this information.
Catversion bumped, initdb required.
Windows), arrange for each postmaster child process to be its own process
group leader, and deliver signals SIGINT, SIGTERM, SIGQUIT to the whole
process group not only the direct child process. This provides saner behavior
for archive and recovery scripts; in particular, it's possible to shut down a
warm-standby recovery server using "pg_ctl stop -m immediate", since delivery
of SIGQUIT to the startup subprocess will result in killing the waiting
recovery_command. Also, this makes Query Cancel and statement_timeout apply
to scripts being run from backends via system(). (There is no support in the
core backend for that, but it's widely done using untrusted PLs.) Per gripe
from Stephen Harris and subsequent discussion.