/*
* DeleteAllExportedSnapshotFiles
* Clean up any files that have been left behind by a crashed backend
* that had exported snapshots before it died.
*
* This should be called during database startup or crash recovery.
*/
void
DeleteAllExportedSnapshotFiles(void)
{
char buf[MAXPGPATH];
DIR *s_dir;
struct dirent *s_de;
if (!(s_dir = AllocateDir(SNAPSHOT_EXPORT_DIR)))
{
/*
* We really should have that directory in a sane cluster setup. But
* then again if we don't, it's not fatal enough to make it FATAL.
* Since we're running in the postmaster, LOG is our best bet.
*/
elog(LOG, "could not open directory \"%s\": %m", SNAPSHOT_EXPORT_DIR);
return;
}
while ((s_de = ReadDir(s_dir, SNAPSHOT_EXPORT_DIR)) != NULL)
{
if (strcmp(s_de->d_name, ".") == 0 ||
strcmp(s_de->d_name, "..") == 0)
continue;
snprintf(buf, MAXPGPATH, SNAPSHOT_EXPORT_DIR "/%s", s_de->d_name);
/* Again, unlink failure is not worthy of FATAL */
if (unlink(buf))
elog(LOG, "could not unlink file \"%s\": %m", buf);
}
FreeDir(s_dir);
}
/*
* pg_start_backup: set up for taking an on-line backup dump
*
* Essentially what this does is to create a backup label file in $PGDATA,
* where it will be archived as part of the backup dump. The label file
* contains the user-supplied label string (typically this would be used
* to tell where the backup dump will be stored) and the starting time and
* starting WAL location for the dump.
*/
Datum
pg_start_backup(PG_FUNCTION_ARGS)
{
text *backupid = PG_GETARG_TEXT_P(0);
bool fast = PG_GETARG_BOOL(1);
char *backupidstr;
XLogRecPtr startpoint;
DIR *dir;
backupidstr = text_to_cstring(backupid);
if (!superuser() && !has_rolreplication(GetUserId()))
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("must be superuser or replication role to run a backup")));
/* Make sure we can open the directory with tablespaces in it */
dir = AllocateDir("pg_tblspc");
if (!dir)
ereport(ERROR,
(errmsg("could not open directory \"%s\": %m", "pg_tblspc")));
startpoint = do_pg_start_backup(backupidstr, fast, NULL, NULL,
dir, NULL, NULL, false, true);
FreeDir(dir);
PG_RETURN_LSN(startpoint);
}
/* Return physical size of directory contents, or 0 if dir doesn't exist */
static int64
db_dir_size(const char *path)
{
int64 dirsize = 0;
struct dirent *direntry;
DIR *dirdesc;
char filename[MAXPGPATH];
dirdesc = AllocateDir(path);
if (!dirdesc)
return 0;
while ((direntry = ReadDir(dirdesc, path)) != NULL)
{
struct stat fst;
if (strcmp(direntry->d_name, ".") == 0 ||
strcmp(direntry->d_name, "..") == 0)
continue;
snprintf(filename, MAXPGPATH, "%s/%s", path, direntry->d_name);
if (stat(filename, &fst) < 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not stat file \"%s\": %m", filename)));
dirsize += fst.st_size;
}
FreeDir(dirdesc);
return dirsize;
}
/*
* Scan the SimpleLRU directory and apply a callback to each file found in it.
*
* If the callback returns true, the scan is stopped. The last return value
* from the callback is returned.
*
* Note that the ordering in which the directory is scanned is not guaranteed.
*
* Note that no locking is applied.
*/
bool
SlruScanDirectory(SlruCtl ctl, SlruScanCallback callback, void *data)
{
bool retval = false;
DIR *cldir;
struct dirent *clde;
int segno;
int segpage;
cldir = AllocateDir(ctl->Dir);
while ((clde = ReadDir(cldir, ctl->Dir)) != NULL)
{
if (strlen(clde->d_name) == 4 &&
strspn(clde->d_name, "0123456789ABCDEF") == 4)
{
segno = (int) strtol(clde->d_name, NULL, 16);
segpage = segno * SLRU_PAGES_PER_SEGMENT;
elog(DEBUG2, "SlruScanDirectory invoking callback on %s/%s",
ctl->Dir, clde->d_name);
retval = callback(ctl, clde->d_name, segpage, data);
if (retval)
break;
}
}
FreeDir(cldir);
return retval;
}
Datum
database_size(PG_FUNCTION_ARGS)
{
Name dbname = PG_GETARG_NAME(0);
Oid dbid;
char *dbpath;
DIR *dirdesc;
struct dirent *direntry;
int64 totalsize;
dbid = get_database_oid(NameStr(*dbname));
if (!OidIsValid(dbid))
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_DATABASE),
errmsg("database \"%s\" does not exist", NameStr(*dbname))));
dbpath = GetDatabasePath(dbid);
dirdesc = AllocateDir(dbpath);
if (!dirdesc)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open directory \"%s\": %m", dbpath)));
totalsize = 0;
for (;;)
{
char *fullname;
struct stat statbuf;
errno = 0;
direntry = readdir(dirdesc);
if (!direntry)
{
if (errno)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("error reading directory: %m")));
else
break;
}
fullname = psnprintf(strlen(dbpath) + 1 + strlen(direntry->d_name) + 1,
"%s/%s", dbpath, direntry->d_name);
if (stat(fullname, &statbuf) == -1)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not stat \"%s\": %m", fullname)));
totalsize += statbuf.st_size;
pfree(fullname);
}
FreeDir(dirdesc);
PG_RETURN_INT64(totalsize);
}
/*
* List a directory (returns the filenames only)
*/
Datum
pg_ls_dir(PG_FUNCTION_ARGS)
{
FuncCallContext *funcctx;
struct dirent *de;
directory_fctx *fctx;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to get directory listings"))));
if (SRF_IS_FIRSTCALL())
{
MemoryContext oldcontext;
funcctx = SRF_FIRSTCALL_INIT();
oldcontext = MemoryContextSwitchTo(funcctx->multi_call_memory_ctx);
fctx = palloc(sizeof(directory_fctx));
fctx->location = convert_and_check_filename(PG_GETARG_TEXT_P(0));
fctx->dirdesc = AllocateDir(fctx->location);
if (!fctx->dirdesc)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open directory \"%s\": %m",
fctx->location)));
funcctx->user_fctx = fctx;
MemoryContextSwitchTo(oldcontext);
}
funcctx = SRF_PERCALL_SETUP();
fctx = (directory_fctx *) funcctx->user_fctx;
while ((de = ReadDir(fctx->dirdesc, fctx->location)) != NULL)
{
int len = strlen(de->d_name);
text *result;
if (strcmp(de->d_name, ".") == 0 ||
strcmp(de->d_name, "..") == 0)
continue;
result = palloc(len + VARHDRSZ);
SET_VARSIZE(result, len + VARHDRSZ);
memcpy(VARDATA(result), de->d_name, len);
SRF_RETURN_NEXT(funcctx, PointerGetDatum(result));
}
FreeDir(fctx->dirdesc);
SRF_RETURN_DONE(funcctx);
}
/*
* calculate total size of tablespace
*/
static int64
calculate_tablespace_size(Oid tblspcOid)
{
char tblspcPath[MAXPGPATH];
char pathname[MAXPGPATH];
int64 totalsize = 0;
DIR *dirdesc;
struct dirent *direntry;
if (tblspcOid == DEFAULTTABLESPACE_OID)
snprintf(tblspcPath, MAXPGPATH, "base");
else if (tblspcOid == GLOBALTABLESPACE_OID)
snprintf(tblspcPath, MAXPGPATH, "global");
else
snprintf(tblspcPath, MAXPGPATH, "pg_tblspc/%u", tblspcOid);
dirdesc = AllocateDir(tblspcPath);
if (!dirdesc)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open tablespace directory \"%s\": %m",
tblspcPath)));
while ((direntry = ReadDir(dirdesc, tblspcPath)) != NULL)
{
struct stat fst;
CHECK_FOR_INTERRUPTS();
if (strcmp(direntry->d_name, ".") == 0 ||
strcmp(direntry->d_name, "..") == 0)
continue;
snprintf(pathname, MAXPGPATH, "%s/%s", tblspcPath, direntry->d_name);
if (stat(pathname, &fst) < 0)
{
if (errno == ENOENT)
continue;
else
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not stat file \"%s\": %m", pathname)));
}
if (S_ISDIR(fst.st_mode))
totalsize += db_dir_size(pathname);
totalsize += fst.st_size;
}
FreeDir(dirdesc);
return totalsize;
}
/*
* CopyTaskFilesFromDirectory finds all files in the given directory, except for
* those having an attempt suffix. The function then copies these files into the
* database table identified by the given schema and table name.
*/
static void
CopyTaskFilesFromDirectory(StringInfo schemaName, StringInfo relationName,
StringInfo sourceDirectoryName)
{
const char *directoryName = sourceDirectoryName->data;
struct dirent *directoryEntry = NULL;
uint64 copiedRowTotal = 0;
DIR *directory = AllocateDir(directoryName);
if (directory == NULL)
{
ereport(ERROR, (errcode_for_file_access(),
errmsg("could not open directory \"%s\": %m", directoryName)));
}
directoryEntry = ReadDir(directory, directoryName);
for (; directoryEntry != NULL; directoryEntry = ReadDir(directory, directoryName))
{
const char *baseFilename = directoryEntry->d_name;
const char *queryString = NULL;
StringInfo fullFilename = NULL;
RangeVar *relation = NULL;
CopyStmt *copyStatement = NULL;
uint64 copiedRowCount = 0;
/* if system file or lingering task file, skip it */
if (strncmp(baseFilename, ".", MAXPGPATH) == 0 ||
strncmp(baseFilename, "..", MAXPGPATH) == 0 ||
strstr(baseFilename, ATTEMPT_FILE_SUFFIX) != NULL)
{
continue;
}
fullFilename = makeStringInfo();
appendStringInfo(fullFilename, "%s/%s", directoryName, baseFilename);
/* build relation object and copy statement */
relation = makeRangeVar(schemaName->data, relationName->data, -1);
copyStatement = CopyStatement(relation, fullFilename->data);
if (BinaryWorkerCopyFormat)
{
DefElem *copyOption = makeDefElem("format", (Node *) makeString("binary"));
copyStatement->options = list_make1(copyOption);
}
DoCopy(copyStatement, queryString, &copiedRowCount);
copiedRowTotal += copiedRowCount;
CommandCounterIncrement();
}
ereport(DEBUG2, (errmsg("copied " UINT64_FORMAT " rows into table: \"%s.%s\"",
copiedRowTotal, schemaName->data, relationName->data)));
FreeDir(directory);
}
/*
* calculate size of database in all tablespaces
*/
static int64
calculate_database_size(Oid dbOid)
{
int64 totalsize;
DIR *dirdesc;
struct dirent *direntry;
char dirpath[MAXPGPATH];
char pathname[MAXPGPATH + 12 + sizeof(TABLESPACE_VERSION_DIRECTORY)];
AclResult aclresult;
/*
* User must have connect privilege for target database
* or be a member of pg_read_all_stats
*/
aclresult = pg_database_aclcheck(dbOid, GetUserId(), ACL_CONNECT);
if (aclresult != ACLCHECK_OK &&
!is_member_of_role(GetUserId(), DEFAULT_ROLE_READ_ALL_STATS))
{
aclcheck_error(aclresult, ACL_KIND_DATABASE,
get_database_name(dbOid));
}
/* Shared storage in pg_global is not counted */
/* Include pg_default storage */
snprintf(pathname, sizeof(pathname), "base/%u", dbOid);
totalsize = db_dir_size(pathname);
/* Scan the non-default tablespaces */
snprintf(dirpath, MAXPGPATH, "pg_tblspc");
dirdesc = AllocateDir(dirpath);
if (!dirdesc)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open tablespace directory \"%s\": %m",
dirpath)));
while ((direntry = ReadDir(dirdesc, dirpath)) != NULL)
{
CHECK_FOR_INTERRUPTS();
if (strcmp(direntry->d_name, ".") == 0 ||
strcmp(direntry->d_name, "..") == 0)
continue;
snprintf(pathname, sizeof(pathname), "pg_tblspc/%s/%s/%u",
direntry->d_name, TABLESPACE_VERSION_DIRECTORY, dbOid);
totalsize += db_dir_size(pathname);
}
FreeDir(dirdesc);
return totalsize;
}
/*
* pgarch_readyXlog
*
* Return name of the oldest xlog file that has not yet been archived.
* No notification is set that file archiving is now in progress, so
* this would need to be extended if multiple concurrent archival
* tasks were created. If a failure occurs, we will completely
* re-copy the file at the next available opportunity.
*
* It is important that we return the oldest, so that we archive xlogs
* in order that they were written, for two reasons:
* 1) to maintain the sequential chain of xlogs required for recovery
* 2) because the oldest ones will sooner become candidates for
* recycling at time of checkpoint
*
* NOTE: the "oldest" comparison will presently consider all segments of
* a timeline with a smaller ID to be older than all segments of a timeline
* with a larger ID; the net result being that past timelines are given
* higher priority for archiving. This seems okay, or at least not
* obviously worth changing.
*/
static bool
pgarch_readyXlog(char *xlog)
{
/*
* open xlog status directory and read through list of xlogs that have the
* .ready suffix, looking for earliest file. It is possible to optimise
* this code, though only a single file is expected on the vast majority
* of calls, so....
*/
char XLogArchiveStatusDir[MAXPGPATH];
char newxlog[MAX_XFN_CHARS + 6 + 1];
DIR *rldir;
struct dirent *rlde;
bool found = false;
snprintf(XLogArchiveStatusDir, MAXPGPATH, XLOGDIR "/archive_status");
rldir = AllocateDir(XLogArchiveStatusDir);
if (rldir == NULL)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open archive status directory \"%s\": %m",
XLogArchiveStatusDir)));
while ((rlde = ReadDir(rldir, XLogArchiveStatusDir)) != NULL)
{
int basenamelen = (int) strlen(rlde->d_name) - 6;
if (basenamelen >= MIN_XFN_CHARS &&
basenamelen <= MAX_XFN_CHARS &&
strspn(rlde->d_name, VALID_XFN_CHARS) >= basenamelen &&
strcmp(rlde->d_name + basenamelen, ".ready") == 0)
{
if (!found)
{
strcpy(newxlog, rlde->d_name);
found = true;
}
else
{
if (strcmp(rlde->d_name, newxlog) < 0)
strcpy(newxlog, rlde->d_name);
}
}
}
FreeDir(rldir);
if (found)
{
/* truncate off the .ready */
newxlog[strlen(newxlog) - 6] = '\0';
strcpy(xlog, newxlog);
}
return found;
}
/*
* calculate size of database in all tablespaces
*/
static int64
calculate_database_size(Oid dbOid)
{
int64 totalsize;
DIR *dirdesc;
struct dirent *direntry;
char dirpath[MAXPGPATH];
char pathname[MAXPGPATH];
AclResult aclresult;
/* User must have connect privilege for target database */
aclresult = pg_database_aclcheck(dbOid, GetUserId(), ACL_CONNECT);
if (aclresult != ACLCHECK_OK)
aclcheck_error(aclresult, ACL_KIND_DATABASE,
get_database_name(dbOid));
/* Shared storage in pg_global is not counted */
/* Include pg_default storage */
snprintf(pathname, MAXPGPATH, "base/%u", dbOid);
totalsize = db_dir_size(pathname);
/* Scan the non-default tablespaces */
snprintf(dirpath, MAXPGPATH, "pg_tblspc");
dirdesc = AllocateDir(dirpath);
if (!dirdesc)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open tablespace directory \"%s\": %m",
dirpath)));
while ((direntry = ReadDir(dirdesc, dirpath)) != NULL)
{
CHECK_FOR_INTERRUPTS();
if (strcmp(direntry->d_name, ".") == 0 ||
strcmp(direntry->d_name, "..") == 0)
continue;
snprintf(pathname, MAXPGPATH, "pg_tblspc/%s/%s/%u",
direntry->d_name, TABLESPACE_VERSION_DIRECTORY, dbOid);
totalsize += db_dir_size(pathname);
}
FreeDir(dirdesc);
/* Complain if we found no trace of the DB at all */
if (!totalsize)
ereport(ERROR,
(ERRCODE_UNDEFINED_DATABASE,
errmsg("database with OID %u does not exist", dbOid)));
return totalsize;
}
/*
* Reset unlogged relations from before the last restart.
*
* If op includes UNLOGGED_RELATION_CLEANUP, we remove all forks of any
* relation with an "init" fork, except for the "init" fork itself.
*
* If op includes UNLOGGED_RELATION_INIT, we copy the "init" fork to the main
* fork.
*/
void
ResetUnloggedRelations(int op)
{
char temp_path[MAXPGPATH];
DIR *spc_dir;
struct dirent *spc_de;
MemoryContext tmpctx,
oldctx;
/* Log it. */
elog(DEBUG1, "resetting unlogged relations: cleanup %d init %d",
(op & UNLOGGED_RELATION_CLEANUP) != 0,
(op & UNLOGGED_RELATION_INIT) != 0);
/*
* Just to be sure we don't leak any memory, let's create a temporary
* memory context for this operation.
*/
tmpctx = AllocSetContextCreate(CurrentMemoryContext,
"ResetUnloggedRelations",
ALLOCSET_DEFAULT_MINSIZE,
ALLOCSET_DEFAULT_INITSIZE,
ALLOCSET_DEFAULT_MAXSIZE);
oldctx = MemoryContextSwitchTo(tmpctx);
/*
* First process unlogged files in pg_default ($PGDATA/base)
*/
ResetUnloggedRelationsInTablespaceDir("base", op);
/*
* Cycle through directories for all non-default tablespaces.
*/
spc_dir = AllocateDir("pg_tblspc");
while ((spc_de = ReadDir(spc_dir, "pg_tblspc")) != NULL)
{
if (strcmp(spc_de->d_name, ".") == 0 ||
strcmp(spc_de->d_name, "..") == 0)
continue;
snprintf(temp_path, sizeof(temp_path), "pg_tblspc/%s/%s",
spc_de->d_name, TABLESPACE_VERSION_DIRECTORY);
ResetUnloggedRelationsInTablespaceDir(temp_path, op);
}
FreeDir(spc_dir);
/*
* Restore memory context.
*/
MemoryContextSwitchTo(oldctx);
MemoryContextDelete(tmpctx);
}
/*
* Load all replication slots from disk into memory at server startup. This
* needs to be run before we start crash recovery.
*/
void
StartupReplicationSlots(XLogRecPtr checkPointRedo)
{
DIR *replication_dir;
struct dirent *replication_de;
ereport(DEBUG1,
(errmsg("starting up replication slots")));
/* restore all slots by iterating over all on-disk entries */
replication_dir = AllocateDir("pg_replslot");
while ((replication_de = ReadDir(replication_dir, "pg_replslot")) != NULL)
{
struct stat statbuf;
char path[MAXPGPATH];
if (strcmp(replication_de->d_name, ".") == 0 ||
strcmp(replication_de->d_name, "..") == 0)
continue;
snprintf(path, MAXPGPATH, "pg_replslot/%s", replication_de->d_name);
/* we're only creating directories here, skip if it's not our's */
if (lstat(path, &statbuf) == 0 && !S_ISDIR(statbuf.st_mode))
continue;
/* we crashed while a slot was being setup or deleted, clean up */
if (string_endswith(replication_de->d_name, ".tmp"))
{
if (!rmtree(path, true))
{
ereport(WARNING,
(errcode_for_file_access(),
errmsg("could not remove directory \"%s\"", path)));
continue;
}
fsync_fname("pg_replslot", true);
continue;
}
/* looks like a slot in a normal state, restore */
RestoreSlotFromDisk(replication_de->d_name);
}
FreeDir(replication_dir);
/* currently no slots exist, we're done. */
if (max_replication_slots <= 0)
return;
/* Now that we have recovered all the data, compute replication xmin */
ReplicationSlotsComputeRequiredXmin();
ReplicationSlotsComputeRequiredLSN();
}
/*
* pg_start_backup: set up for taking an on-line backup dump
*
* Essentially what this does is to create a backup label file in $PGDATA,
* where it will be archived as part of the backup dump. The label file
* contains the user-supplied label string (typically this would be used
* to tell where the backup dump will be stored) and the starting time and
* starting WAL location for the dump.
*
* Permission checking for this function is managed through the normal
* GRANT system.
*/
Datum
pg_start_backup(PG_FUNCTION_ARGS)
{
text *backupid = PG_GETARG_TEXT_P(0);
bool fast = PG_GETARG_BOOL(1);
bool exclusive = PG_GETARG_BOOL(2);
char *backupidstr;
XLogRecPtr startpoint;
DIR *dir;
backupidstr = text_to_cstring(backupid);
if (exclusive_backup_running || nonexclusive_backup_running)
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("a backup is already in progress in this session")));
/* Make sure we can open the directory with tablespaces in it */
dir = AllocateDir("pg_tblspc");
if (!dir)
ereport(ERROR,
(errmsg("could not open directory \"%s\": %m", "pg_tblspc")));
if (exclusive)
{
startpoint = do_pg_start_backup(backupidstr, fast, NULL, NULL,
dir, NULL, NULL, false, true);
exclusive_backup_running = true;
}
else
{
MemoryContext oldcontext;
/*
* Label file and tablespace map file need to be long-lived, since they
* are read in pg_stop_backup.
*/
oldcontext = MemoryContextSwitchTo(TopMemoryContext);
label_file = makeStringInfo();
tblspc_map_file = makeStringInfo();
MemoryContextSwitchTo(oldcontext);
startpoint = do_pg_start_backup(backupidstr, fast, NULL, label_file,
dir, NULL, tblspc_map_file, false, true);
nonexclusive_backup_running = true;
before_shmem_exit(nonexclusive_base_backup_cleanup, (Datum) 0);
}
FreeDir(dir);
PG_RETURN_LSN(startpoint);
}
/*
* copydir: copy a directory
*
* If recurse is false, subdirectories are ignored. Anything that's not
* a directory or a regular file is ignored.
*/
void
copydir(char *fromdir, char *todir, bool recurse)
{
DIR *xldir;
struct dirent *xlde;
char fromfile[MAXPGPATH];
char tofile[MAXPGPATH];
if (mkdir(todir, S_IRUSR | S_IWUSR | S_IXUSR) != 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not create directory \"%s\": %m", todir)));
xldir = AllocateDir(fromdir);
if (xldir == NULL)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open directory \"%s\": %m", fromdir)));
while ((xlde = ReadDir(xldir, fromdir)) != NULL)
{
struct stat fst;
/* If we got a cancel signal during the copy of the directory, quit */
CHECK_FOR_INTERRUPTS();
if (strcmp(xlde->d_name, ".") == 0 ||
strcmp(xlde->d_name, "..") == 0)
continue;
snprintf(fromfile, MAXPGPATH, "%s/%s", fromdir, xlde->d_name);
snprintf(tofile, MAXPGPATH, "%s/%s", todir, xlde->d_name);
if (lstat(fromfile, &fst) < 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not stat file \"%s\": %m", fromfile)));
if (S_ISDIR(fst.st_mode))
{
/* recurse to handle subdirectories */
if (recurse)
copydir(fromfile, tofile, true);
}
else if (S_ISREG(fst.st_mode))
copy_file(fromfile, tofile);
}
FreeDir(xldir);
}
static int
SlruRecoverMirrorDir(char *slru_dir)
{
DIR *cldir;
struct dirent *clde;
int retval = 0;
char *dir = NULL;
char *mirrorDir = NULL;
int counter = 0;
if (isTxnDir(slru_dir))
{
dir = makeRelativeToTxnFilespace(slru_dir);
mirrorDir = makeRelativeToPeerTxnFilespace(slru_dir);
}
else
{
dir = (char*)palloc(MAXPGPATH);
strncpy(dir, slru_dir, MAXPGPATH);
mirrorDir = (char*)palloc(MAXPGPATH);
strncpy(mirrorDir, slru_dir, MAXPGPATH);
}
cldir = AllocateDir(dir);
while ((clde = ReadDir(cldir, dir)) != NULL) {
if (strlen(clde->d_name) == 4 &&
strspn(clde->d_name, "0123456789ABCDEF") == 4) {
retval = MirrorFlatFile(slru_dir, clde->d_name);
if (retval != 0)
break;
}
counter++;
if (counter % log_count_recovered_files_batch == 0)
{
elog(LOG, "completed recovering %d files", counter);
}
}
elog(LOG, "completed recovering %d files", counter);
FreeDir(cldir);
pfree(dir);
pfree(mirrorDir);
return retval;
}
/*
* calculate total size of tablespace
*/
Datum
pg_tablespace_size(PG_FUNCTION_ARGS)
{
Oid tblspcOid = PG_GETARG_OID(0);
char tblspcPath[MAXPGPATH];
char pathname[MAXPGPATH];
int64 totalsize=0;
DIR *dirdesc;
struct dirent *direntry;
if (tblspcOid == DEFAULTTABLESPACE_OID)
snprintf(tblspcPath, MAXPGPATH, "%s/base", DataDir);
else if (tblspcOid == GLOBALTABLESPACE_OID)
snprintf(tblspcPath, MAXPGPATH, "%s/global", DataDir);
else
snprintf(tblspcPath, MAXPGPATH, "%s/pg_tblspc/%u", DataDir, tblspcOid);
dirdesc = AllocateDir(tblspcPath);
if (!dirdesc)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open tablespace directory \"%s\": %m",
tblspcPath)));
while ((direntry = readdir(dirdesc)) != NULL)
{
struct stat fst;
if (strcmp(direntry->d_name, ".") == 0 ||
strcmp(direntry->d_name, "..") == 0)
continue;
snprintf(pathname, MAXPGPATH, "%s/%s", tblspcPath, direntry->d_name);
if (stat(pathname, &fst) < 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not stat \"%s\": %m", pathname)));
totalsize += fst.st_size;
if (fst.st_mode & S_IFDIR)
totalsize += db_dir_size(pathname);
}
FreeDir(dirdesc);
PG_RETURN_INT64(totalsize);
}
pg_tzenum *
pg_tzenumerate_start(void)
{
pg_tzenum *ret = (pg_tzenum *) palloc0(sizeof(pg_tzenum));
char *startdir = pstrdup(pg_TZDIR());
ret->baselen = strlen(startdir) + 1;
ret->depth = 0;
ret->dirname[0] = startdir;
ret->dirdesc[0] = AllocateDir(startdir);
if (!ret->dirdesc[0])
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open directory \"%s\": %m", startdir)));
return ret;
}
/*
* calculate size of database in all tablespaces
*/
static int64
calculate_database_size(Oid dbOid)
{
int64 totalsize;
DIR *dirdesc;
struct dirent *direntry;
char dirpath[MAXPGPATH];
char pathname[MAXPGPATH];
/* Shared storage in pg_global is not counted */
/* Include pg_default storage */
snprintf(pathname, MAXPGPATH, "base/%u", dbOid);
totalsize = db_dir_size(pathname);
/* Scan the non-default tablespaces */
snprintf(dirpath, MAXPGPATH, "pg_tblspc");
dirdesc = AllocateDir(dirpath);
if (!dirdesc)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open tablespace directory \"%s\": %m",
dirpath)));
while ((direntry = ReadDir(dirdesc, dirpath)) != NULL)
{
CHECK_FOR_INTERRUPTS();
if (strcmp(direntry->d_name, ".") == 0 ||
strcmp(direntry->d_name, "..") == 0)
continue;
snprintf(pathname, MAXPGPATH, "pg_tblspc/%s/%u",
direntry->d_name, dbOid);
totalsize += db_dir_size(pathname);
}
FreeDir(dirdesc);
/* Complain if we found no trace of the DB at all */
if (!totalsize)
ereport(ERROR,
(ERRCODE_UNDEFINED_DATABASE,
errmsg("database with OID %u does not exist", dbOid)));
return totalsize;
}
/* Process one pgsql_tmp directory for RemovePgTempFiles */
static void
RemovePgTempFilesInDir(const char *tmpdirname)
{
DIR *temp_dir;
struct dirent *temp_de;
char rm_path[MAXPGPATH];
temp_dir = AllocateDir(tmpdirname);
if (temp_dir == NULL)
{
/* anything except ENOENT is fishy */
if (errno != ENOENT)
elog(LOG,
"could not open temporary-files directory \"%s\": %m",
tmpdirname);
return;
}
while ((temp_de = ReadDir(temp_dir, tmpdirname)) != NULL)
{
if (strcmp(temp_de->d_name, ".") == 0 ||
strcmp(temp_de->d_name, "..") == 0)
continue;
snprintf(rm_path, sizeof(rm_path), "%s/%s",
tmpdirname, temp_de->d_name);
if (HasTempFilePrefix(temp_de->d_name))
{
/*
* It can be a file or a directory, so try to delete both ways
* We ignore errors.
*/
unlink(rm_path);
rmtree(rm_path,true);
}
else
{
elog(LOG,
"unexpected file found in temporary-files directory: \"%s\"",
rm_path);
}
}
FreeDir(temp_dir);
}
/* Return physical size of directory contents, or 0 if dir doesn't exist */
int64
db_dir_size(const char *path)
{
int64 dirsize = 0;
struct dirent *direntry;
char filename[MAXPGPATH];
/* Deal with remote shared storage */
if (!IsLocalPath(path))
{
dirsize += HdfsPathSize(path);
}
/* Local storage */
else
{
DIR *dirdesc = AllocateDir(path);
if (!dirdesc)
return 0;
while ((direntry = ReadDir(dirdesc, path)) != NULL)
{
struct stat fst;
if (strcmp(direntry->d_name, ".") == 0 ||
strcmp(direntry->d_name, "..") == 0)
continue;
snprintf(filename, MAXPGPATH, "%s/%s", path, direntry->d_name);
if (stat(filename, &fst) < 0)
{
if (errno == ENOENT)
continue;
else
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not stat file \"%s\": %m", filename)));
}
dirsize += fst.st_size;
}
FreeDir(dirdesc);
}
return dirsize;
}
/*
* Process one tablespace directory for ResetUnloggedRelations
*/
static void
ResetUnloggedRelationsInTablespaceDir(const char *tsdirname, int op)
{
DIR *ts_dir;
struct dirent *de;
char dbspace_path[MAXPGPATH * 2];
ts_dir = AllocateDir(tsdirname);
/*
* If we get ENOENT on a tablespace directory, log it and return. This
* can happen if a previous DROP TABLESPACE crashed between removing the
* tablespace directory and removing the symlink in pg_tblspc. We don't
* really want to prevent database startup in that scenario, so let it
* pass instead. Any other type of error will be reported by ReadDir
* (causing a startup failure).
*/
if (ts_dir == NULL && errno == ENOENT)
{
ereport(LOG,
(errcode_for_file_access(),
errmsg("could not open directory \"%s\": %m",
tsdirname)));
return;
}
while ((de = ReadDir(ts_dir, tsdirname)) != NULL)
{
/*
* We're only interested in the per-database directories, which have
* numeric names. Note that this code will also (properly) ignore "."
* and "..".
*/
if (strspn(de->d_name, "0123456789") != strlen(de->d_name))
continue;
snprintf(dbspace_path, sizeof(dbspace_path), "%s/%s",
tsdirname, de->d_name);
ResetUnloggedRelationsInDbspaceDir(dbspace_path, op);
}
FreeDir(ts_dir);
}
/*
* SimpleLruTruncate subroutine: scan directory for removable segments.
* Actually remove them iff doDeletions is true. Return TRUE iff any
* removable segments were found. Note: no locking is needed.
*
* This can be called directly from clog.c, for reasons explained there.
*/
bool
SlruScanDirectory(SlruCtl ctl, int cutoffPage, bool doDeletions)
{
bool found = false;
DIR *cldir;
struct dirent *clde;
int segno;
int segpage;
char path[MAXPGPATH];
/*
* The cutoff point is the start of the segment containing cutoffPage.
* (This is redundant when called from SimpleLruTruncate, but not when
* called directly from clog.c.)
*/
cutoffPage -= cutoffPage % SLRU_PAGES_PER_SEGMENT;
cldir = AllocateDir(ctl->Dir);
while ((clde = ReadDir(cldir, ctl->Dir)) != NULL)
{
if (strlen(clde->d_name) == 4 &&
strspn(clde->d_name, "0123456789ABCDEF") == 4)
{
segno = (int) strtol(clde->d_name, NULL, 16);
segpage = segno * SLRU_PAGES_PER_SEGMENT;
if (ctl->PagePrecedes(segpage, cutoffPage))
{
found = true;
if (doDeletions)
{
snprintf(path, MAXPGPATH, "%s/%s", ctl->Dir, clde->d_name);
ereport(DEBUG2,
(errmsg("removing file \"%s\"", path)));
unlink(path);
}
}
}
}
FreeDir(cldir);
return found;
}
/*
* Check if a directory is empty.
*
* This probably belongs somewhere else, but not sure where...
*/
bool
directory_is_empty(const char *path)
{
DIR *dirdesc;
struct dirent *de;
dirdesc = AllocateDir(path);
while ((de = ReadDir(dirdesc, path)) != NULL)
{
if (strcmp(de->d_name, ".") == 0 ||
strcmp(de->d_name, "..") == 0)
continue;
FreeDir(dirdesc);
return false;
}
FreeDir(dirdesc);
return true;
}
/*
* copydir: copy a directory (we only need to go one level deep)
*
* Return 0 on success, nonzero on failure.
*
* NB: do not elog(ERROR) on failure. Return to caller so it can try to
* clean up.
*/
int
copydir(char *fromdir, char *todir)
{
DIR *xldir;
struct dirent *xlde;
char fromfl[MAXPGPATH];
char tofl[MAXPGPATH];
if (mkdir(todir) != 0)
{
ereport(WARNING,
(errcode_for_file_access(),
errmsg("could not create directory \"%s\": %m", todir)));
return -1;
}
xldir = AllocateDir(fromdir);
if (xldir == NULL)
{
ereport(WARNING,
(errcode_for_file_access(),
errmsg("could not open directory \"%s\": %m", fromdir)));
return -1;
}
while ((xlde = readdir(xldir)) != NULL)
{
snprintf(fromfl, MAXPGPATH, "%s/%s", fromdir, xlde->d_name);
snprintf(tofl, MAXPGPATH, "%s/%s", todir, xlde->d_name);
if (CopyFile(fromfl, tofl, TRUE) < 0)
{
ereport(WARNING,
(errcode_for_file_access(),
errmsg("could not copy file \"%s\": %m", fromfl)));
FreeDir(xldir);
return -1;
}
}
FreeDir(xldir);
return 0;
}
/*
* Scan specified directory for a case-insensitive match to fname
* (of length fnamelen --- fname may not be null terminated!). If found,
* copy the actual filename into canonname and return true.
*/
static bool
scan_directory_ci(const char *dirname, const char *fname, int fnamelen,
char *canonname, int canonnamelen)
{
bool found = false;
DIR *dirdesc;
struct dirent *direntry;
dirdesc = AllocateDir(dirname);
if (!dirdesc)
{
ereport(LOG,
(errcode_for_file_access(),
errmsg("could not open directory \"%s\": %m", dirname)));
return false;
}
while ((direntry = ReadDir(dirdesc, dirname)) != NULL)
{
/*
* Ignore . and .., plus any other "hidden" files. This is a security
* measure to prevent access to files outside the timezone directory.
*/
if (direntry->d_name[0] == '.')
continue;
if (strlen(direntry->d_name) == fnamelen &&
pg_strncasecmp(direntry->d_name, fname, fnamelen) == 0)
{
/* Found our match */
strlcpy(canonname, direntry->d_name, canonnamelen);
found = true;
break;
}
}
FreeDir(dirdesc);
return found;
}
/*
* Remove temporary files left over from a prior postmaster session
*
* This should be called during postmaster startup. It will forcibly
* remove any leftover files created by OpenTemporaryFile.
*
* NOTE: we could, but don't, call this during a post-backend-crash restart
* cycle. The argument for not doing it is that someone might want to examine
* the temp files for debugging purposes. This does however mean that
* OpenTemporaryFile had better allow for collision with an existing temp
* file name.
*/
void
RemovePgTempFiles(void)
{
char temp_path[MAXPGPATH];
DIR *spc_dir;
struct dirent *spc_de;
/*
* First process temp files in pg_default ($PGDATA/base)
*/
snprintf(temp_path, sizeof(temp_path), "base/%s", PG_TEMP_FILES_DIR);
RemovePgTempFilesInDir(temp_path);
/*
* Cycle through temp directories for all non-default tablespaces.
*/
spc_dir = AllocateDir("pg_tblspc");
while ((spc_de = ReadDir(spc_dir, "pg_tblspc")) != NULL)
{
if (strcmp(spc_de->d_name, ".") == 0 ||
strcmp(spc_de->d_name, "..") == 0)
continue;
snprintf(temp_path, sizeof(temp_path), "pg_tblspc/%s/%s",
spc_de->d_name, PG_TEMP_FILES_DIR);
RemovePgTempFilesInDir(temp_path);
}
FreeDir(spc_dir);
/*
* In EXEC_BACKEND case there is a pgsql_tmp directory at the top level of
* DataDir as well.
*/
#ifdef EXEC_BACKEND
RemovePgTempFilesInDir(PG_TEMP_FILES_DIR);
#endif
}
/* Process one pgsql_tmp directory for RemovePgTempFiles */
static void
RemovePgTempFilesInDir(const char *tmpdirname)
{
DIR *temp_dir;
struct dirent *temp_de;
char rm_path[MAXPGPATH];
temp_dir = AllocateDir(tmpdirname);
if (temp_dir == NULL)
{
/* anything except ENOENT is fishy */
if (errno != ENOENT)
elog(LOG,
"could not open temporary-files directory \"%s\": %m",
tmpdirname);
return;
}
while ((temp_de = ReadDir(temp_dir, tmpdirname)) != NULL)
{
if (strcmp(temp_de->d_name, ".") == 0 ||
strcmp(temp_de->d_name, "..") == 0)
continue;
snprintf(rm_path, sizeof(rm_path), "%s/%s",
tmpdirname, temp_de->d_name);
if (strncmp(temp_de->d_name,
PG_TEMP_FILE_PREFIX,
strlen(PG_TEMP_FILE_PREFIX)) == 0)
unlink(rm_path); /* note we ignore any error */
else
elog(LOG,
"unexpected file found in temporary-files directory: \"%s\"",
rm_path);
}
FreeDir(temp_dir);
}
/* Process one tablespace directory for ResetUnloggedRelations */
static void
ResetUnloggedRelationsInTablespaceDir(const char *tsdirname, int op)
{
DIR *ts_dir;
struct dirent *de;
char dbspace_path[MAXPGPATH];
ts_dir = AllocateDir(tsdirname);
if (ts_dir == NULL)
{
/* anything except ENOENT is fishy */
if (errno != ENOENT)
elog(LOG,
"could not open tablespace directory \"%s\": %m",
tsdirname);
return;
}
while ((de = ReadDir(ts_dir, tsdirname)) != NULL)
{
int i = 0;
/*
* We're only interested in the per-database directories, which have
* numeric names. Note that this code will also (properly) ignore "."
* and "..".
*/
while (isdigit((unsigned char) de->d_name[i]))
++i;
if (de->d_name[i] != '\0' || i == 0)
continue;
snprintf(dbspace_path, sizeof(dbspace_path), "%s/%s",
tsdirname, de->d_name);
ResetUnloggedRelationsInDbspaceDir(dbspace_path, op);
}
FreeDir(ts_dir);
}
static bool cfs_gc_directory(int worker_id, char const* path)
{
DIR* dir = AllocateDir(path);
bool success = true;
if (dir != NULL) {
struct dirent* entry;
char file_path[MAXPGPATH];
int len;
while ((entry = ReadDir(dir, path)) != NULL && !cfs_stop)
{
if (strcmp(entry->d_name, ".") == 0 ||
strcmp(entry->d_name, "..") == 0)
{
continue;
}
len = snprintf(file_path, sizeof(file_path), "%s/%s", path, entry->d_name);
if (len > 4 &&
strcmp(file_path + len - 4, ".map") == 0)
{
if (entry->d_ino % cfs_state->n_workers == worker_id && !cfs_gc_file(file_path))
{
success = false;
break;
}
} else {
if (!cfs_gc_directory(worker_id, file_path))
{
success = false;
break;
}
}
}
FreeDir(dir);
}
return success;
}