MySQL replace into with null

I am using replace into to update one table with the information in another. The first table is called srv_update and it includes:

srv_id,    , srv_list_price,   srv_list_price
'120', 'Third Feline PCR', '20.25'
'121', 'Flu Rhino Vacc', '26.00'
'122', 'Scaly Mite', '35.00'
'123', 'Intestinal Parasite Screen', '26.00'
'124', 'Tick Removal', '15.00'
'125', 'Behaviour Modification', '75.00'
'126', 'Vitamin E- Concentrated', '30.00'
'127', 'Sedative-Feline', '25.00'
'128', 'Flea Treatment- Small Animal', '35.00'
'129', 'Flea Treatment- Large Animal', '50.00'

the second table is srv_update_changes_2 and it includes:

item_id,   item_desc,      item_list_price
'134', 'Vitamin E series', '55.00'
'135', 'Feline PCR Series', '75.00'
'114', 'H1N1 vaccine', '75.00'
'115', 'H1N2 vaccine', '75.00'
'127', 'Sedative- small mammal', '75.00'
'111', 'Rabies_V-Rodent', '25.00'
'129', NULL, '25.00'

I want to only replace existing IDs and not insert any new ID's. Also if the desc field is null (it is for id 129 of the second table) when replacing the null field should keep the value of the original table (the descripton in the first table)

129 is the only row that should be affected. I added it to the table to try to figure this out but I cant seem to get the logic. Ive tried ifnull and ive tried coalesce but i cant seem to get the right output. it either stays null in the updated table or inserts the wrong description.

Ive also attempted coalesce where the second field is a subquery that checks if the first field is null to replace it with the service description from the first table where that rows item id matched the srv id in the second table where its desc row is null

replace into upd_services
select item_id,
coalesce(item_desc,(
    select srv_desc 
from upd_services 
where srv_id in(
    select item_id
    from upd_services_changes_2
    where item_desc is null))) srv_desc,
item_list_price
from upd_services_changes_2 
where item_id in (
select srv_id
from upd_services)

thanks for any help

Solution

I would use update with a join:

update upd_services us join
       upd_services_changes_2 usc
       on us.srv_id = usc.item_id
    set us.srv_desc = coalesce(usc.item_desc, us.srv_desc),
        us.srv_list_price = usc.item_list_price;