update rebuild instructions

This commit is contained in:
Matt 2015-01-06 13:06:42 -08:00
parent 24fd6a1a26
commit c307cce330
3 changed files with 42 additions and 6 deletions

2
Conf.h
View File

@ -829,7 +829,7 @@ class Conf {
char m_rebuildRoots ;
char m_rebuildNonRoots ;
char m_rebuildSkipSitedbLookup ;
//char m_rebuildSkipSitedbLookup ;
// for caching the qualities of urls (see Msg20.cpp)
int32_t m_maxQualityCacheAge ;

View File

@ -18406,7 +18406,13 @@ void Parms::init ( ) {
m->m_title = "recycle link info";
m->m_desc = "If enabled, gigablast will recycle the link info "
"when rebuilding titledb.";
"when rebuilding titledb. The link info is basically just "
"the link text. The siterank, which is determined by the "
"number of inlinks to a site, is stored/cached in tagdb "
"so that is a separate item. If you want to pick up new "
"link text you will want to set this to <i>NO</i> and "
"make sure to rebuild titledb, since that stores the "
"link text.";
m->m_cgi = "rrli"; // repair full rebuild
m->m_off = (char *)&g_conf.m_rebuildRecycleLinkInfo - g;
m->m_type = TYPE_BOOL;
@ -18637,10 +18643,13 @@ void Parms::init ( ) {
m->m_group = 0;
m++;
/*
m->m_title = "skip tagdb lookup";
m->m_desc = "When rebuilding spiderdb and scanning it for new spiderdb "
"records, should a tagdb lookup be performed? Runs much much "
"faster without it. Will also keep the original doc quality and "
m->m_desc = "When rebuilding spiderdb and scanning it for new "
"spiderdb records, should a tagdb lookup be performed? "
"Runs much much "
"faster without it. Will also keep the original doc quality "
"and "
"spider priority in tact.";
m->m_cgi = "rssl";
m->m_off = (char *)&g_conf.m_rebuildSkipSitedbLookup - g;
@ -18650,6 +18659,7 @@ void Parms::init ( ) {
m->m_def = "0";
m->m_group = 0;
m++;
*/
///////////////////////////////////////////
// END PAGE REPAIR //

View File

@ -2292,7 +2292,33 @@ bool Repair::printRepairStatus ( SafeBuf *sb , int32_t fromIp ) {
"<tr bgcolor=#%s><td colspan=2>"
"<font size=-2>"
"Use this to rebuild a database or to reindex "
"all pages to pick up new link text."
"all pages to pick up new link text. Or to "
"reindex all pages to pick up new site rank info "
"from tagdb. To pick up "
"new link text you should rebuild titledb and posdb. "
"If unsure, just do a full rebuild, but it will "
"require about 2GB more than the disk used before "
"the rebuild, so at its peak the rebuild will use "
"a little more than double the disk space you "
"are using now. Also you will want to set "
"recycle link info to false to pick up the new link "
"text. However, if you just want to pick up "
"new sitenuminlinks tags in tagdb to get more "
"accurate siteranks for each result, then you can "
"leave the recycle link info set to true."
""
"<br><br>"
"All spidering for all collections will be disabled "
"when the rebuild is in progress. But you should "
"still be able to conduct searches on the original "
"index. You can pause "
"the rebuild by disabling <i>rebuild mode enabled"
"</i>. Each shard should save its rebuid state so "
"you can safely shut shards down when rebuilding "
"and they should resume on startup. When the rebuild "
"completes it moves the original files to the trash "
"subdirectory and replaces them with the newly "
"rebuilt files."
"</font>"
"</td></tr>"