FTL/src/gc.c

308 lines
9.1 KiB
C

/* Pi-hole: A black hole for Internet advertisements
* (c) 2017 Pi-hole, LLC (https://pi-hole.net)
* Network-wide ad blocking via your own hardware.
*
* FTL Engine
* Garbage collection routines
*
* This file is copyright under the latest version of the EUPL.
* Please see LICENSE file for your rights under this license. */
#include "FTL.h"
#include "gc.h"
#include "shmem.h"
#include "timers.h"
#include "config.h"
#include "overTime.h"
#include "database/common.h"
#include "log.h"
// global variable killed
#include "signals.h"
// data getter functions
#include "datastructure.h"
// logg_rate_limit_message()
#include "database/message-table.h"
// get_nprocs()
#include <sys/sysinfo.h>
// get_filepath_usage()
#include "files.h"
// Resource checking interval
// default: 300 seconds
#define RCinterval 300
bool doGC = false;
// Subtract rate-limitation count from individual client counters
// As long as client->rate_limit is still larger than the allowed
// maximum count, the rate-limitation will just continue
static void reset_rate_limiting(void)
{
for(int clientID = 0; clientID < counters->clients; clientID++)
{
clientsData *client = getClient(clientID, true);
if(!client)
continue;
// Check if we are currently rate-limiting this client
if(client->flags.rate_limited)
{
const char *clientIP = getstr(client->ippos);
// Check if we want to continue rate limiting
if(client->rate_limit > config.rate_limit.count)
{
logg("Still rate-limiting %s as it made additional %d queries", clientIP, client->rate_limit);
}
// or if rate-limiting ends for this client now
else
{
logg("Ending rate-limitation of %s", clientIP);
client->flags.rate_limited = false;
}
}
// Reset counter
client->rate_limit = 0;
}
}
static time_t lastRateLimitCleaner = 0;
// Returns how many more seconds until the current rate-limiting interval is over
time_t get_rate_limit_turnaround(const unsigned int rate_limit_count)
{
const unsigned int how_often = rate_limit_count/config.rate_limit.count;
return (time_t)config.rate_limit.interval*how_often - (time(NULL) - lastRateLimitCleaner);
}
static int check_space(const char *file, int LastUsage)
{
if(config.check.disk == 0)
return 0;
int perc = 0;
char buffer[64] = { 0 };
// Warn if space usage at the device holding the corresponding file
// exceeds the configured threshold and current usage is higher than
// usage in the last run (to prevent log spam)
perc = get_filepath_usage(file, buffer);
if(perc > config.check.disk && perc > LastUsage && perc <= 100.0)
log_resource_shortage(-1.0, 0, -1, perc, file, buffer);
return perc;
}
static void check_load(void)
{
if(!config.check.load)
return;
// Get CPU load averages
double load[3];
if (getloadavg(load, 3) == -1)
return;
// Get number of CPU cores
const int nprocs = get_nprocs();
// Warn if 15 minute average of load exceeds number of available
// processors
if(load[2] > nprocs)
log_resource_shortage(load[2], nprocs, -1, -1, NULL, NULL);
}
void *GC_thread(void *val)
{
// Set thread name
thread_names[GC] = "housekeeper";
prctl(PR_SET_NAME, thread_names[GC], 0, 0, 0);
// Remember when we last ran the actions
time_t lastGCrun = time(NULL) - time(NULL)%GCinterval;
lastRateLimitCleaner = time(NULL);
time_t lastResourceCheck = 0;
// Remember disk usage
int LastLogStorageUsage = 0;
int LastDBStorageUsage = 0;
// Run as long as this thread is not canceled
while(!killed)
{
const time_t now = time(NULL);
if((unsigned int)(now - lastRateLimitCleaner) >= config.rate_limit.interval)
{
lastRateLimitCleaner = now;
lock_shm();
reset_rate_limiting();
unlock_shm();
}
// Intermediate cancellation-point
if(killed)
break;
// Check available resources
if(now - lastResourceCheck >= RCinterval)
{
check_load();
LastDBStorageUsage = check_space(FTLfiles.FTL_db, LastDBStorageUsage);
LastLogStorageUsage = check_space(FTLfiles.log, LastLogStorageUsage);
lastResourceCheck = now;
}
if(now - GCdelay - lastGCrun >= GCinterval || doGC)
{
doGC = false;
// Update lastGCrun timer
lastGCrun = now - GCdelay - (now - GCdelay)%GCinterval;
// Lock FTL's data structure, since it is likely that it will be changed here
// Requests should not be processed/answered when data is about to change
lock_shm();
// Get minimum timestamp to keep (this can be set with MAXLOGAGE)
time_t mintime = (now - GCdelay) - config.maxlogage;
// Align the start time of this GC run to the GCinterval. This will also align with the
// oldest overTime interval after GC is done.
mintime -= mintime % GCinterval;
if(config.debug & DEBUG_GC)
{
timer_start(GC_TIMER);
char timestring[84] = "";
get_timestr(timestring, mintime, false);
logg("GC starting, mintime: %s (%llu)", timestring, (long long)mintime);
}
// Process all queries
int removed = 0;
for(long int i=0; i < counters->queries; i++)
{
queriesData* query = getQuery(i, true);
if(query == NULL)
continue;
// Test if this query is too new
if(query->timestamp > mintime)
break;
// Adjust client counter (total and overTime)
clientsData* client = getClient(query->clientID, true);
const int timeidx = getOverTimeID(query->timestamp);
overTime[timeidx].total--;
if(client != NULL)
change_clientcount(client, -1, 0, timeidx, -1);
// Adjust domain counter (no overTime information)
domainsData* domain = getDomain(query->domainID, true);
if(domain != NULL)
domain->count--;
// Get upstream pointer
// Change other counters according to status of this query
switch(query->status)
{
case QUERY_UNKNOWN:
// Unknown (?)
break;
case QUERY_FORWARDED: // (fall through)
case QUERY_RETRIED: // (fall through)
case QUERY_RETRIED_DNSSEC:
// Forwarded to an upstream DNS server
// Adjusting counters is done below in moveOverTimeMemory()
break;
case QUERY_CACHE:
case QUERY_CACHE_STALE:
// Answered from local cache _or_ local config
break;
case QUERY_GRAVITY: // Blocked by Pi-hole's blocking lists (fall through)
case QUERY_BLACKLIST: // Exact blocked (fall through)
case QUERY_REGEX: // Regex blocked (fall through)
case QUERY_EXTERNAL_BLOCKED_IP: // Blocked by upstream provider (fall through)
case QUERY_EXTERNAL_BLOCKED_NXRA: // Blocked by upstream provider (fall through)
case QUERY_EXTERNAL_BLOCKED_NULL: // Blocked by upstream provider (fall through)
case QUERY_GRAVITY_CNAME: // Gravity domain in CNAME chain (fall through)
case QUERY_REGEX_CNAME: // Regex blacklisted domain in CNAME chain (fall through)
case QUERY_BLACKLIST_CNAME: // Exactly blacklisted domain in CNAME chain (fall through)
case QUERY_DBBUSY: // Blocked because gravity database was busy
case QUERY_SPECIAL_DOMAIN: // Blocked by special domain handling
if(domain != NULL)
domain->blockedcount--;
if(client != NULL)
change_clientcount(client, 0, -1, -1, 0);
break;
case QUERY_IN_PROGRESS: // Don't have to do anything here
case QUERY_STATUS_MAX: // fall through
default:
/* That cannot happen */
break;
}
// Update reply counters
counters->reply[query->reply]--;
// Update type counters
if(query->type >= TYPE_A && query->type < TYPE_MAX)
{
counters->querytype[query->type-1]--;
}
// Set query again to UNKNOWN to reset the counters
query_set_status(query, QUERY_UNKNOWN);
// Finally, remove the last trace of this query
counters->status[QUERY_UNKNOWN]--;
// Count removed queries
removed++;
}
// Only perform memory operations when we actually removed queries
if(removed > 0)
{
// Move memory forward to keep only what we want
// Note: for overlapping memory blocks, memmove() is a safer approach than memcpy()
// Example: (I = now invalid, X = still valid queries, F = free space)
// Before: IIIIIIXXXXFF
// After: XXXXFFFFFFFF
queriesData *dest = getQuery(0, true);
queriesData *src = getQuery(removed, true);
if(dest && src)
memmove(dest, src, (counters->queries - removed)*sizeof(queriesData));
// Update queries counter
counters->queries -= removed;
// Update DB index as total number of queries reduced
lastdbindex -= removed;
// ensure remaining memory is zeroed out (marked as "F" in the above example)
queriesData *tail = getQuery(counters->queries, true);
if(tail)
memset(tail, 0, (counters->queries_MAX - counters->queries)*sizeof(queriesData));
}
// Determine if overTime memory needs to get moved
moveOverTimeMemory(mintime);
if(config.debug & DEBUG_GC)
logg("Notice: GC removed %i queries (took %.2f ms)", removed, timer_elapsed_msec(GC_TIMER));
// Release thread lock
unlock_shm();
// After storing data in the database for the next time,
// we should scan for old entries, which will then be deleted
// to free up pages in the database and prevent it from growing
// ever larger and larger
DBdeleteoldqueries = true;
}
thread_sleepms(GC, 1000);
}
logg("Terminating GC thread");
return NULL;
}