Difference between docs stored and indexed

Elastic Stack Elasticsearch
1

Hi,

looking at my _stats I can find the following informations:

_all: {
primaries: {
docs: {
count: 8617,
deleted: 47
},
indexing: {
index_total: 9017,
index_time: "2.3s",
index_time_in_millis: 2308,
index_current: 0,
delete_total: 0,
delete_time: "0s",
delete_time_in_millis: 0,
delete_current: 0

In my database I have 9017 objects like the value for indexing.index_total
So what does this difference of 400 docs to docs.count mean? I cannot find
them in my search results or facet searches. Was there a problem during the
import? Should I recreate the index? Where can I find additional
information?

Thanks,
Konstantin

--

2

Hello Konstantin,

If you index two documents with the same ID, they will be counted twice in
"index_total" and only once in docs count. That's because you indexed two
docs and only have one stored.

Also, if you delete the document, you'll end up with 0 docs, but still 2 in
index_total.

Best regards,
Radu

http://sematext.com/ -- Elasticsearch -- Solr -- Lucene

On Wed, Nov 28, 2012 at 7:01 PM, Konstantin Krauss <
konstantin@propertybase.com> wrote:

Hi,

looking at my _stats I can find the following informations:

_all: {
primaries: {
docs: {
count: 8617,
deleted: 47
},
indexing: {
index_total: 9017,
index_time: "2.3s",
index_time_in_millis: 2308,
index_current: 0,
delete_total: 0,
delete_time: "0s",
delete_time_in_millis: 0,
delete_current: 0

In my database I have 9017 objects like the value for indexing.index_total
So what does this difference of 400 docs to docs.count mean? I cannot find
them in my search results or facet searches. Was there a problem during the
import? Should I recreate the index? Where can I find additional
information?

Thanks,
Konstantin

--

3

Hi Radu,

thanks for the explanation. This makes "index_total" much clearer.
So I have to investigate, why some docs are getting deleted. Duplicate IDs
cannot be the case, because I take the IDs from my database and the index
was new.

Best,
Konstantin

--