yacy_search_server/yacy.init

###
### YACY Init File
###
# These properties will be loaded upon installation.
# They are used only once for set-up.
# If you make changes to this file and want these to make any effect,
# you must delete the httpProxy.conf file in DATA/SETTINGS

# ----------------------------------------------------------------------------
# the http service configurations

# port number where the server should bind to
# e.g. 8080
#      #eth0:8080
# 	   192.168.0.1:8080
port = 8080

# time-out of client control socket in milliseconds
# since this applies only to the client-proxy connection,
# it can be rather short
# milliseconds
clientTimeout = 8000

# maximal number of httpd sessions
# a client may open several connections at one, and the httpdMaxActiveSessions value sets
# a limit on the number of concurrent connections
httpdMaxActiveSessions = 150
httpdMaxIdleSessions = 75
httpdMinIdleSessions = 5

# default root path for the file server
# may be overridden by the htdocs parameter
# users shall be encouraged to use the htdocs path for individual content,
# not this path defined here
htRootPath = htroot
htTemplatePath = htroot/env/templates

# individual htroot folder
# every user may publicise her/his own web pages
# these pages shall be placed in the path defined here
# the htdocs path shares its content with the htroot path
htDocsPath = DATA/HTDOCS

# the default files (typically index.html), if no file name is given
# The complete path to this file is created by combination with the rootPath
# you can set a list of defaults, separated by comma
# the first one is priorized
defaultFiles = index.html,default.html,search.html,console.html,control.html,welcome.html,wiki.html,forum.html,blog.html,email.html,content.html,monitor.html,share.html,dir.html,readme.txt

# locale-options: yacy supports localization.
# Web pages for special languages are located in the htLocalePath
# The htLocaleLang defines a list of language options as <dir>/<named-language>
# the <dir> must exist as sub-path to htLocalePath
# the htLocaleSelection selects from the given locales, value=one-of-<dir>
htDefaultPath=htroot
htLocalePath=htroot/locale
htLocaleLang=default/English,de/Deutsch,fr/Fran&ccedil;ais,nl/Nederlands,it/Italiano,es/Espa&ntilde;ol,pt/Portug&ecirc;s,fi/Suomi,se/Svenska,dk/Dansk,gr/E&lambda;&lambda;&eta;v&iota;&kappa;&alpha;
htLocaleSelection=default

# virtual host for httpdFileServlet access
# for example http://<fileHost>/ shall access the file servlet and
# return the defaultFile at rootPath
# either way, http://<fileHost>/ denotes the same as http://localhost:<port>/
# for the preconfigured value 'localpeer', the URL is:
# http://localpeer/
fileHost = localpeer

# root path for message files
messPath = C:/AnomicServer

# specify the path to the MIME matching file table
mimeConfig = httpd.mime

# specify the path to message resource file
messConfig = httpd.messages

# proxy use. This server can also act as an caching proxy.
# to enable that function, set proxy=true
proxy=true

# a path to the proxy's file cache.
# This will be used if the server is addressed as a proxy
proxyCache = DATA/HTCACHE

# the proxy's maximum disc cache size in megabytes
# there should be enough space for the browsing load of an internet caffee
# running at 56kbit/s modem speed (this time not unusual)
# during 3 days, 8 hours a day
# necessary space = 3 * 8 * 60 * 60 * 56 / 8 = 604800 KB = ca. 590 MB
# since 600 MB is not much these days (it's below one GB!)
# we recommend using that space
#proxyCacheSize = 600
#for testing:
proxyCacheSize = 200

# the following mime-types are the whitelist for indexing
#
# parseableRealtimeMimeTypes: specifies mime-types that can be indexed on the fly
# parseableMime: specifies mime-types that can be indexed but not on the fly
parseableRealtimeMimeTypes=application/xhtml+xml,text/html,text/plain
parseableMimeTypes=

# media extension string
# a comma-separated list of extensions that denote media file formats
# this is important to recognize <a href> - tags as not-html reference
# These files will be excluded from indexing _(Please keep extensions in alphabetical order)_
mediaExt=ace,arj,asf,avi,bin,bz2,css,deb,doc,dll,dmg,gif,gz,hqx,ico,img,iso,jar,jpe,jpg,jpeg,mpeg,mov,mp3,mpg,ogg,png,pdf,ppt,ps,ram,rar,rm,rpm,sit,so,swf,sxc,sxd,sxi,sxw,tar,tgz,torrent,wmv,xcf,xls,zip
parseableExt=html,htm,txt,php,shtml,asp,aspx,jsp

# Promotion Strings
# These strings appear in the Web Mask of the YACY search client
# Set these Strings to cusomize your peer and give any message to
# other peer users
promoteSearchPageGreeting =

# the path to the PLASMA database, especially the reverse word index
dbPath=DATA/PLASMADB

# the path to the LISTS files. Most lists are used to filter web content
listsPath=DATA/LISTS

# the path to the SKINS files.
skinPath=DATA/SKINS

# the yellow-list; URL's elements
# (the core of an URL; like 'yahoo' in 'de.yahoo.com')
# appearing in this list will not get a manipulated user agent string
proxyYellowList=yacy.yellow

# the black-list; URLs appearing in this list will not be loaded;
# instead always a 404 is returned
# all these files will be placed in the listsPath
proxyBlackLists=url.default.black
proxyBlackListsActive=url.default.black
proxyBlackListsShared=url.default.black
proxyCookieBlackList=cookie.default.black
proxyCookieWhiteList=cookie.default.black

# the blue-list; 
# no search result is locally presented that has any word of the bluelist
# in the search words, the URL or the URL's description
plasmaBlueList=yacy.blue

# this proxy may in turn again access another proxy
# if you wish to do that, specify it here
# if you want to switch on the proxy use, set remoteProxyUse=true
# remoteProxyNoProxy is a no-proxy pattern list for the remote proxy
remoteProxyUse=false
remoteProxyUse4Yacy=true
remoteProxyUse4SSL=true

remoteProxyHost=192.168.2.2
remoteProxyPort=4239
remoteProxyUser=
remoteProxyPwd=

remoteProxyNoProxy=192.*,10.*,127.*,localhost

# the proxy may filter the content of transferred web pages
# the bluelist removes specific keywords from web pages
proxyBlueList=yacy.blue

# security settigns
# we provide proxy and server security through a 2-stage security gate:
# 1st stage: firewall-like access control trough ip filter for clients
# 2nd stage: password settings for proxy, server and server administrators
# by default, these settings are weak to simplify set-up and testing
# every user/adiministrator shall be encouraged to change these settings
# your can change them also online during run-time on
# http://localhost:8080/

# proxyClient: client-ip's that may connect the proxy for proxy service
# if several ip's are allowed then they must be separated by a ','
# any ip may contain the wildcard-sign '*'
#proxyClient=192.168.0.4
proxyClient=localhost,127.0.0.1,192.168.*,10.*

# serverClient: client-ip's that may connect to the web server,
# thus are allowed to use the search service
# if you set this to another value, search requst from others
# are blocked, but you will also be blocked from using others
# search services.
serverClient=*

### proxyAccount: a user:password - pair for proxy authentification
### leave empty for no authenication
### example:
##proxyAccount=jim:knopf
##proxyAccount=
##proxyAccountBase64MD5=

# use_proxyAccounts: set to true to restrict proxy-access to some identified users.
#use User_p.html to create some Users.
use_proxyAccounts=false

# serverAccount: a user:password - pair for web server access
# this is the access to the 'public' pages on the server
# should be always open, but you get the option here
# if set to a user:password, you get a conflict with the administration account
# future versions will check if the server is unprotected,
# because the p2p-index-sharing function will use the http server for
# data exchange.
# example
#serverAccount=dicke:berta
serverAccount=
serverAccountBase64MD5=

# adminAccount: a user:password - pair for administration of
# settings through the web interface
# should be set to a secret. By default it is without a password
# but you are encouraged to set it to another value on the page
# http://localhost:8080/
#adminAccount=admin:anomic
adminAccount=
adminAccountBase64MD5=

# peer-to-peer construction for distributed search
# we have several stages:
# 1st: a file within every distribution that has a list of URLs:
#      -> this is the superseed file
# 2nd: the files that can be retrieved by the superseeds' URLs
#      are called seed list-files.
#      -> the seed list-files contain IP/port combinations of running
#         AnomicHTTPProxies
# 3rd: the peers that are targeted within the seed files are called superpeers
# 4th: the superpeers hold and share a list of all client/search/crawl peers
#
# some superpeers should be able to create again seed list-files.
# These superpeers must upload their IP or their list of peer-IP's to a
# ftp location to create the seed list-file.
# Everyone who do so should mail his/her new seed location to mc<at>anomic.de
# The seed list-file location will then be included in the superseed file.
# This superseed file is available then at two localtions:
# - it is included in every distribution and
# - updated through a specific URL-location
# we see the file name and the URL of the superseed here:
superseedFile=superseed.txt
superseedLocation=http://www.yacy.net/superseed.txt


# if you are running a principal peer, you must update the following variables
# The upload method that should be used to upload the seed-list file to
# a public accessible webserver where it can be loaded by other peers.
#
# You can set the seedUploadMethod-Property to 
# - None
# - Ftp
# - File
# - Scp (only if you have installed the optional addon)
#
seedUploadMethod=

# The URL to the seed list file 
seedURL=

# This is the most common method to upload the seed-list
#  
# This is an ftp account with all relevant information.
# The update is only made if there had been changes in between.
seedFTPServer=
seedFTPAccount=
seedFTPPassword=
seedFTPPath=

# alternatively to an FTP account, a peer can also become a principal peer
# if the seed-list can be generated as a file and that file is also accessible from
# the internet. In this case, omit any ftp settings and set this path here.
# if this path stays empty, an ftp account is considered
# however, you must always set a seedURL because it is used to check if the
# file is actually accessible from the internet
seedFilePath=

# Settings needed to upload the seed-list file via scp
#
# Please note that this upload method can only be used if you have installed
# this optional upload method.
seedScpServer=
seedScpAccount=
seedScpPassword=
seedScpPath=

# every peer should have a name. inded, we try to give every peer an unique ID,
# which is necessary for internal organization of the index sharing, but the
# peer's name is purely informal. No function but information is applied.
# please change this at your pleasure
peerName=anomic

# every peer periodically scans for other peers. you can set the time
# of the period here (minutes)
peerCycle=2

# The p2p maintenance can run in either of two online modes:
# - don't process jobs and only access available in cache -> mode 0
# - process any job only if we are online, which is technically only the case
#   if the proxy is used -> mode 1
# - process jobs periodically, with periodes according to peerCycle -> mode 2
#onlineMode=1
onlineMode=2

# Debug mode for YACY network: this will trigger that also local ip's are
# accepted as peer addresses
yacyDebugMode=false

#staticIP if you have a static IP, you can use this setting
staticIP=

# if the process is running behind a NAT or ROUTER, we cannot easily identify
# the public IP of the process. We can ask a public IP responder, but cannot
# rely on it. Therefore, AnomicHTTPProxy includes it's own responder.
# But for the first running peer this is not an option.
# The author uses a DI-604 router, which can be
# asked for the public IP. If you own a DI-604 as well, please set the
# DI604use to true and put in your router password, it will not be used for any
# other purpose of asking for the IP
#DI604use=true
DI604use=false
DI604pw=

# each time the proxy starts up, it can trigger the local browser to show the
# status page. This is active by default, to make it easier for first-time
# users to understand what this application does. You can disable browser
# pop-up here or set a different start page, like the search page
# the browser type is optional and works only under certain conditions
#browserPopUpTrigger=false
browserPopUpTrigger=true
#browserPopUpPage=index.html
browserPopUpPage=Status.html
browserPopUpApplication=netscape

# the proxy saves it's own seed information. It is positive for the network if
# the seed does not change it's configuration often (or not at all).
# The reason for that is that the seed hash is the target for the
# distributed hash table distribution function.
# The following file will contain the saved seed:
yacyOwnSeedFile=DATA/YACYDB/mySeed.txt
yacyDB=DATA/YACYDB

# index sharing attributes
# by default, sharing is on. If you want to use the proxy only for
# local indexing, you may switch this off
allowDistributeIndex=true
allowDistributeIndexWhileCrawling=false
allowReceiveIndex=true
indexReceiveBlockBlacklist=false

# the frequency is the number of links per minute, that the peer allowes
# _every_ other peer to send to this peer
defaultWordReceiveFrequency=100
defaultLinkReceiveFrequency=30
# the default may be overridden for each peer individually, these
# settings are only available through the online interface

# prefetch parameters
# the prefetch depth assigns a specific depth to the prefetch mechanism
# prefetch of 0 means no prefetch; a prefetch of 1 means to prefetch all
# embedded URLs, but since embedded image links are loaded by the browser
# this means that only embedded anchors are prefetched additionally
# a prefetch of 2 would result in loading of all images and anchor pages
# of all embedded anchors. Be careful with this value, since even a prefetch
# of 2 would result in hundreds of prefetched URLs for each single proxy fill.
proxyPrefetchDepth=0
proxyStoreHTCache=true

# From the 'IndexCreate' menu point you can also define a crawling start point.
# The crawling works the same way as the prefetch, but it is possible to
# assign a different crawling depth.
# Be careful with this number. Consider a branching factor of average 20;
# A prefect-depth of 8 would index 25.600.000.000 pages, maybe the whole WWW.
crawlingDepth=2
localIndexing=true

# Filter for crawlinig; may be used to restrict a crawl to a specific domain
# URLs are only indexed and further crawled if they match this filter
crawlingFilter=.*
crawlingQ=false
storeHTCache=false
storeTXCache=true

# default crawl profile entries
# if these entries are empty, then a new entry will be generated
defaultProxyProfile=
defaultRemoteProfile=

# peers may initiate remote crawling tasks.
# every peer may allow or disallow to be used as crawling-peer;
# you can also set a maximum crawl depth that can be requested or accepted
# order=parameters for requester; response=parameters for responder
# these values apply only for senior-senior - communication
# The delay value is number of seconds bewteen two separate orders
crawlOrder=true
crawlOrderDepth=0
crawlOrderDelay=8
crawlResponse=true
crawlResponseDepth=0

# indexing-exclusion - rules
# There rules are important to reduce the number of words that are indexed
# We distinguish three different sets of stop-words:
# static   - excludes all words given in the file yacy.stopwords from indexing,
# dynamic  - excludes all words from indexing which are listed by statistic rules,
# parental - excludes all words from indexing which had been indexed in the parent web page.
xsstopw=true
xdstopw=true
xpstopw=true

# performance-settings
# delay-times for permanent loops (milliseconds)
# the idlesleep is the pause that an proces sleeps if the last call to the
# process job was without execution of anything;
# the busysleep is the pause after a full job execution
# the prereq-value is a memory pre-requisite: that much bytes must
# be available/free in the heap; othervise the loop is not executed
# and another idlesleep is performed
20_dhtdistribution_idlesleep=20000
20_dhtdistribution_busysleep=5000
20_dhtdistribution_memprereq=8388608
30_peerping_idlesleep=120000
30_peerping_busysleep=120000
30_peerping_memprereq=1048576
40_peerseedcycle_idlesleep=1800000
40_peerseedcycle_busysleep=1200000
40_peerseedcycle_memprereq=4194304
50_localcrawl_idlesleep=10000
50_localcrawl_busysleep=200
50_localcrawl_memprereq=1048576
61_globalcrawltrigger_idlesleep=10000
61_globalcrawltrigger_busysleep=200
61_globalcrawltrigger_memprereq=1048576
62_remotetriggeredcrawl_idlesleep=10000
62_remotetriggeredcrawl_busysleep=200
62_remotetriggeredcrawl_memprereq=1048576
70_cachemanager_idlesleep=5000
70_cachemanager_busysleep=0
70_cachemanager_memprereq=1048576
80_indexing_idlesleep=5000
80_indexing_busysleep=0
80_indexing_memprereq=2097152
82_crawlstack_idlesleep=5000
82_crawlstack_busysleep=0
82_crawlstack_memprereq=1048576
90_cleanup_idlesleep=300000
90_cleanup_busysleep=300000
90_cleanup_memprereq=0

# multiprocessor-settings
# you may want to run time-consuming processes on several processors
# the most time-consuming process is the indexing-Process
# We implemented an option to run several of these processes here
# setting the number of processes to Zero is not allowed
# If you have a double-processor system,
# a cluster value of '2' would be appropriate
80_indexing_cluster=1

# ram cache for database files

# ram cache for assortment cache cluster (for all 64 files)
ramCacheRWI     =  8388608

# ram cache for responseHeader.db
ramCacheHTTP    =  4194304

# ram cache for urlHash.db
ramCacheLURL    =  4194304

# ram cache for urlNotice.db
ramCacheNURL    =  4194304

# ram cache for urlErr.db 
ramCacheEURL    =     8192

# ram cache for seedDBs
ramCacheDHT     =     8192

# ram cache for message.db
ramCacheMessage =     8192

# ram cache for wiki.db
ramCacheWiki    =     8192

# ram cache for news1.db
ramCacheNews    =     8192

# ram cache for robotsTxt.db
ramCacheRobots  =  2097152

# ram cache for crawlProfile.db
ramCacheProfiles  =   8192

# ram cache for stack crawl thread db
ramCachePreNURL = 4194304

# default memory settings for startup of yacy
# is only valid in unix/shell environments and
# not for first startup of YaCy

# -Xmx<size> set maximum Java heap size
javastart_Xmx=Xmx64m

# -Xms<size> set initial Java heap size
javastart_Xms=Xms10m


# performance properties for the word index cache
# wordCacheMaxLow/High is the number of word indexes that shall be held in the
# ram cache during indexing. When YaCy is shut down, this cache must be
# flushed to disc; this may last some minutes.
# The low value is valid for crawling tasks, the high value is valid for
# remote index transmissions and search requests
# maxWaitingWordFlush gives the number of seconds that the shutdown
# may last for the word flush
wordCacheMaxLow  = 12000
wordCacheMaxHigh = 16000
maxWaitingWordFlush = 180

# Specifies if yacy can be used as transparent http proxy.
# 
# Please note that you also have to reconfigure your firewall
# before you can use yacy as transparent proxy. On linux this
# can be done like this:
#   iptables -t nat -A PREROUTING -p tcp -s 192.168.0.0/16 \
#   --dport 80 -j DNAT --to 192.168.0.1:8080
#
# With this iptables filter listed above all http traffic that
# comes from your private network (in this case 192.168.0.0)
# and goes to any webserver listening on port 80 will be forwarded
# by the firewall to yacy running on port 8080 (192.168.0.1:8080)
isTransparentProxy=false

# Specifies if yacy should use the http connection keep-alive feature
connectionKeepAliveSupport=true

# Configuration options needed to configure server port forwarding
portForwardingEnabled=false
portForwardingUseProxy=false
portForwardingPort=

portForwardingHost=
portForwardingHostPort=22
portForwardingHostUser=
portForwardingHostPwd=

# msgForwarding: Specifies if yacy should forward received messages via
# email to the configured email address
msgForwardingEnabled=false
msgForwardingCmd=/usr/sbin/sendmail
msgForwardingTo=root@localhost

#onlineCautionDelay: delay time after proxy usage before crawling is resumed
onlineCautionDelay=30000

# Some configuration values for the crawler
crawler.acceptLanguage=en-us,en;q=0.5
crawler.acceptCharset=ISO-8859-1,utf-8;q=0.7,*;q=0.7
crawler.clientTimeout=9000
crawler.isPaused=false

# maximum number of crawler threads
crawler.MaxActiveThreads = 10
crawler.MaxIdleThreads = 7
crawler.MinIdleThreads = 5

# maximum number of crawl-stacker threads
stacker.MaxActiveThreads = 50
stacker.MaxIdleThreads = 10
stacker.MinIdleThreads = 5

# maximum size of indexing queue
indexer.slots = 100

# specifies if yacy should set it's own referer if no referer URL
# was set by the client.
useYacyReferer = true

#allow only 443(https-port) for https-proxy?
#if you want to tunnel other protokols, set to false
secureHttps = true

# specifies if the httpdFileHandler should cache 
# the template-files from the htroot directory
enableTemplateCache = true

# specifies if the http post body should be transfered
# using content-encoding gzip during index transfer
# a) indexDistribution: which is done periodically if you have enabled 
#    Index Distribution via IndexControl_p.html
# b) indexTransfer: which can be used to transfer the whole index of a peer
#    this can be started via IndexTransfer_p.html
# c) indexControl: which can be triggered manually via IndexControl_p.html to
#    transfer a chosen subset of the peer index     
indexDistribution.gzipBody = true
indexTransfer.gzipBody = true
indexControl.gzipBody = true

# defining timeouts for index- transfer/distribution/control
indexControl.timeout = 60000
indexDistribution.timeout = 60000
indexTransfer.timeout = 120000

# defining max. allowed amount of open files during index- transfer/distribution
indexDistribution.maxOpenFiles = 800
indexTransfer.maxOpenFiles = 800

# 
storagePeerHash =

# Search sequence settings
# collection:
# time = time to get a RWI out of RAM cache, assortments and WORDS files
# count = maximum number of RWI-entries that shall be collected
#
# join:
# time = time to perform the join between all collected RWIs
# count = maximum number of entries that shall be joined
#
# presort:
# time = time to do a sort of the joined URL-records
# count = maximum number of entries that shall be pre-sorted
#
# urlfetch:
# time = time to fetch the real URLs from the LURL database
# count = maximum number of urls that shall be fetched
#
# postsort:
# time = time for final sort of URLs
# count = maximum number oof URLs that shall be retrieved during sort
#
# filter:
# time = time to filter out unwanted urls (like redundant urls)
# count = maximum number of urls that shall be filtered
#
# snippetfetch:
# time = time to fetch snippets for selected URLs
# count = maximum number of snipptes to be fetched
#
# all values are percent
# time-percent is the percent of total search time
# count-percent is the percent of total wanted urls in result
# we distinguish local and remote search times
searchProcessLocalTime_c = 25
searchProcessLocalCount_c = 10000000
searchProcessLocalTime_j = 10
searchProcessLocalCount_j = 1000000
searchProcessLocalTime_r = 10
searchProcessLocalCount_r =100000
searchProcessLocalTime_u = 30
searchProcessLocalCount_u = 10000
searchProcessLocalTime_o = 10
searchProcessLocalCount_o = 100
searchProcessLocalTime_f = 5
searchProcessLocalCount_f = 100
searchProcessLocalTime_s = 10
searchProcessLocalCount_s = 30

searchProcessRemoteTime_c =  25
searchProcessRemoteCount_c = 1000000
searchProcessRemoteTime_j = 10
searchProcessRemoteCount_j = 1000000
searchProcessRemoteTime_r = 10 
searchProcessRemoteCount_r = 1000
searchProcessRemoteTime_u = 30
searchProcessRemoteCount_u = 1000
searchProcessRemoteTime_o = 10
searchProcessRemoteCount_o = 1000
searchProcessRemoteTime_f = 5
searchProcessRemoteCount_f = 100
searchProcessRemoteTime_s = 10
searchProcessRemoteCount_s = 10

# path to ranking directory containing ranking reference files
rankingPath = DATA/RANKING

# a list of domain name patterns that should not be cached by the httpc dns cache
httpc.nameCacheNoCachingPatterns = .*.dyndns.org, .*.dynalias.org